View HD6417618_4352868.PDF datasheet online --- IC-ON-LINE

Datasheet File OCR Text:

REJ09B0131-0600
The revision list can be viewed directly by clicking the title page. The revision list summarizes the locations of revisions and additions. Details should always be checked by referring to the relevant text.
32
SuperH
TM
SH7618 Group
Hardware Manual
Renesas 32-Bit RISC Microcomputer RISC engine Family / SH7618 Series SH7618 SH7618A HD6417618 HD6417618A
Rev.6.00 Revision Date: Jun. 12, 2007
Rev. 6.00 Jun. 12, 2007 Page ii of xxxii
Notes regarding these materials
1. This document is provided for reference purposes only so that Renesas customers may select the appropriate Renesas products for their use. Renesas neither makes warranties or representations with respect to the accuracy or completeness of the information contained in this document nor grants any license to any intellectual property rights or any other rights of Renesas or any third party with respect to the information in this document. 2. Renesas shall have no liability for damages or infringement of any intellectual property or other rights arising out of the use of any information in this document, including, but not limited to, product data, diagrams, charts, programs, algorithms, and application circuit examples. 3. You should not use the products or the technology described in this document for the purpose of military applications such as the development of weapons of mass destruction or for the purpose of any other military use. When exporting the products or technology described herein, you should follow the applicable export control laws and regulations, and procedures required by such laws and regulations. 4. All information included in this document such as product data, diagrams, charts, programs, algorithms, and application circuit examples, is current as of the date this document is issued. Such information, however, is subject to change without any prior notice. Before purchasing or using any Renesas products listed in this document, please confirm the latest product information with a Renesas sales office. Also, please pay regular and careful attention to additional and different information to be disclosed by Renesas such as that disclosed through our website. (http://www.renesas.com ) 5. Renesas has used reasonable care in compiling the information included in this document, but Renesas assumes no liability whatsoever for any damages incurred as a result of errors or omissions in the information included in this document. 6. When using or otherwise relying on the information in this document, you should evaluate the information in light of the total system before deciding about the applicability of such information to the intended application. Renesas makes no representations, warranties or guaranties regarding the suitability of its products for any particular application and specifically disclaims any liability arising out of the application and use of the information in this document or Renesas products. 7. With the exception of products specified by Renesas as suitable for automobile applications, Renesas products are not designed, manufactured or tested for applications or otherwise in systems the failure or malfunction of which may cause a direct threat to human life or create a risk of human injury or which require especially high quality and reliability such as safety systems, or equipment or systems for transportation and traffic, healthcare, combustion control, aerospace and aeronautics, nuclear power, or undersea communication transmission. If you are considering the use of our products for such purposes, please contact a Renesas sales office beforehand. Renesas shall have no liability for damages arising out of the uses set forth above. 8. Notwithstanding the preceding paragraph, you should not use Renesas products for the purposes listed below: (1) artificial life support devices or systems (2) surgical implantations (3) healthcare intervention (e.g., excision, administration of medication, etc.) (4) any other purposes that pose a direct threat to human life Renesas shall have no liability for damages arising out of the uses set forth in the above and purchasers who elect to use Renesas products in any of the foregoing applications shall indemnify and hold harmless Renesas Technology Corp., its affiliated companies and their officers, directors, and employees against any and all damages arising out of such applications. 9. You should use the products described herein within the range specified by Renesas, especially with respect to the maximum rating, operating supply voltage range, movement power voltage range, heat radiation characteristics, installation and other product characteristics. Renesas shall have no liability for malfunctions or damages arising out of the use of Renesas products beyond such specified ranges. 10. Although Renesas endeavors to improve the quality and reliability of its products, IC products have specific characteristics such as the occurrence of failure at a certain rate and malfunctions under certain use conditions. Please be sure to implement safety measures to guard against the possibility of physical injury, and injury or damage caused by fire in the event of the failure of a Renesas product, such as safety design for hardware and software including but not limited to redundancy, fire control and malfunction prevention, appropriate treatment for aging degradation or any other applicable measures. Among others, since the evaluation of microcomputer software alone is very difficult, please evaluate the safety of the final products or system manufactured by you. 11. In case Renesas products listed in this document are detached from the products to which the Renesas products are attached or affixed, the risk of accident such as swallowing by infants and small children is very high. You should implement safety measures so that Renesas products may not be easily detached from your products. Renesas shall have no liability for damages arising out of such detachment. 12. This document may not be reproduced or duplicated, in any form, in whole or in part, without prior written approval from Renesas. 13. Please contact a Renesas sales office if you have any questions regarding the information contained in this document, Renesas semiconductor products, or if you have any other inquiries.
Rev. 6.00 Jun. 12, 2007 Page iii of xxxii
Rev. 6.00 Jun. 12, 2007 Page iv of xxxii
General Precautions on Handling of Product
1. Treatment of NC Pins Note: Do not connect anything to the NC pins. The NC (not connected) pins are either not connected to any of the internal circuitry or are used as test pins or to reduce noise. If something is connected to the NC pins, the operation of the LSI is not guaranteed. 2. Treatment of Unused Input Pins Note: Fix all unused input pins to high or low level. Generally, the input pins of CMOS products are high-impedance input pins. If unused pins are in their open states, intermediate levels are induced by noise in the vicinity, a passthrough current flows internally, and a malfunction may occur. 3. Processing before Initialization Note: When power is first supplied, the product's state is undefined. The states of internal circuits are undefined until full power is supplied throughout the chip and a low level is input on the reset pin. During the period where the states are undefined, the register settings and the output state of each pin are also undefined. Design your system so that it does not malfunction because of processing while it is in this undefined state. For those products which have a reset function, reset the LSI immediately after the power supply has been turned on. 4. Prohibition of Access to Undefined or Reserved Addresses Note: Access to undefined or reserved addresses is prohibited. The undefined or reserved addresses may be used to expand functions, or test registers may have been be allocated to these addresses. Do not access these registers; the system's operation is not guaranteed if they are accessed.
Rev. 6.00 Jun. 12, 2007 Page v of xxxii
Configuration of This Manual
This manual comprises the following items: 1. 2. 3. 4. 5. 6. General Precautions on Handling of Product Configuration of This Manual Preface Contents Overview Description of Functional Modules * CPU and System-Control Modules * On-Chip Peripheral Modules The configuration of the functional description of each module differs according to the module. However, the generic style includes the following items: i) Feature ii) Input/Output Pin iii) Register Description iv) Operation v) Usage Note
When designing an application system that includes this LSI, take notes into account. Each section includes notes in relation to the descriptions given, and usage notes are given, as required, as the final part of each section. 7. List of Registers 8. Electrical Characteristics 9. Appendix 10. Main Revisions and Additions in this Edition (only for revised versions) The list of revisions is a summary of points that have been revised or added to earlier versions. This does not include all of the revised contents. For details, see the actual locations in this manual. 11. Index
Rev. 6.00 Jun. 12, 2007 Page vi of xxxii
Preface
The SH7618 Group RISC (Reduced Instruction Set Computer) microcomputers include a Renesas Technology-original RISC CPU as its core, and the peripheral functions required to configure a system. Target Users: This manual was written for users who will use the SH7618 and SH7618A in the design of application systems. Target users are expected to understand the fundamentals of electrical circuits, logical circuits, and microcomputers. Objective: This manual was written to explain the hardware functions and electrical characteristics of the SH7618 and SH7618A to the target users. Refer to the SH-1/SH-2/SH-DSP Software Manual for a detailed description of the instruction set.
Notes on reading this manual: * In order to understand the overall functions of the chip Read the manual according to the contents. This manual can be roughly categorized into parts on the CPU, system control functions, peripheral functions and electrical characteristics. * In order to understand the details of the CPU's functions Read the SH-1/SH-2/SH-DSP Software Manual. * In order to understand the details of a register when its name is known * The addresses, bits, and initial values of the registers are summarized in section 20, List of Registers. Examples: Register name: The following notation is used for cases when the same or a similar function, e.g. 16-bit timer pulse unit or serial communication interface, is implemented on more than one channel: XXX_N (XXX is the register name and N is the channel number) Bit order: The MSB is on the left and the LSB is on the right. Number notation: Binary is B'xxxx, hexadecimal is H'xxxx, decimal is xxxx. Signal notation: An overbar is added to a low-active signal: xxxx Related Manuals: The latest versions of all related manuals are available from our web site. Please ensure you have the latest versions of all documents you require. http://www.renesas.com/
Rev. 6.00 Jun. 12, 2007 Page vii of xxxii
SH7618 Group manuals:
Document Title SH7618 Group Hardware Manual SH-1/SH-2/SH-DSP Software Manual Document No. This manual REJ09B0171
User's manuals for development tools:
Document Title SuperH RISC engine C/C++ Compiler, Assembler, Optimizing Linkage Editor User's Manual SuperH RISC engine High-performance Embedded Workshop 3 User's Manual SuperH RISC engine High-performance Embedded Workshop 3 Tutorial
TM
Document No. REJ10B0152 REJ10B0025 REJ10B0023
Application note:
Document Title SuperH RISC engine C/C++ Compiler Document No. REJ05B0463
All trademarks and registered trademarks are the property of their respective owners.
Rev. 6.00 Jun. 12, 2007 Page viii of xxxii
Rev. 6.00 Jun. 12, 2007 Page ix of xxxii
Contents
Section 1 Overview ................................................................................................. 1
1.1 1.2 1.3 1.4 Features.................................................................................................................................. 2 Block Diagram....................................................................................................................... 6 Pin Assignments .................................................................................................................... 7 Pin Functions ......................................................................................................................... 8
Section 2 CPU ....................................................................................................... 19
2.1 2.2 Features................................................................................................................................ 19 Register Configuration......................................................................................................... 19 2.2.1 General Registers (Rn)............................................................................................ 21 2.2.2 Control Registers .................................................................................................... 21 2.2.3 System Registers..................................................................................................... 22 2.2.4 Initial Values of Registers....................................................................................... 23 Data Formats........................................................................................................................ 24 2.3.1 Register Data Format .............................................................................................. 24 2.3.2 Memory Data Formats ............................................................................................ 24 2.3.3 Immediate Data Formats......................................................................................... 25 Features of Instructions........................................................................................................ 25 2.4.1 RISC Type .............................................................................................................. 25 2.4.2 Addressing Modes .................................................................................................. 28 2.4.3 Instruction Formats ................................................................................................. 31 Instruction Set ...................................................................................................................... 35 2.5.1 Instruction Set by Type........................................................................................... 35 Processing States.................................................................................................................. 47 2.6.1 State Transition....................................................................................................... 47
2.3
2.4
2.5 2.6
Section 3 Cache ..................................................................................................... 49
3.1 Features................................................................................................................................ 49 3.1.1 Cache Structure....................................................................................................... 49 3.1.2 Divided Areas and Cache ....................................................................................... 51 Register Descriptions........................................................................................................... 52 3.2.1 Cache Control Register 1 (CCR1) .......................................................................... 52 3.2.2 Cache Control Register 3 (CCR3) .......................................................................... 53 Operation ............................................................................................................................. 54 3.3.1 Searching Cache ..................................................................................................... 54 3.3.2 Read Access............................................................................................................ 56
3.2
3.3
Rev. 6.00 Jun. 12, 2007 Page x of xxxii
3.4
3.3.3 Write Access ........................................................................................................... 56 3.3.4 Write-Back Buffer .................................................................................................. 57 3.3.5 Coherency of Cache and External Memory ............................................................ 57 Memory-Mapped Cache ...................................................................................................... 58 3.4.1 Address Array ......................................................................................................... 58 3.4.2 Data Array .............................................................................................................. 59 3.4.3 Usage Examples...................................................................................................... 61
Section 4 U Memory..............................................................................................63
4.1 4.2 Features................................................................................................................................ 63 Usage Notes ......................................................................................................................... 63
Section 5 Exception Handling ...............................................................................65
5.1 Overview.............................................................................................................................. 65 5.1.1 Types of Exception Handling and Priority.............................................................. 65 5.1.2 Exception Handling Operations .............................................................................. 66 5.1.3 Exception Handling Vector Table........................................................................... 67 Resets ................................................................................................................................... 69 5.2.1 Types of Resets....................................................................................................... 69 5.2.2 Power-On Reset ...................................................................................................... 69 5.2.3 H-UDI Reset ........................................................................................................... 70 Address Errors ..................................................................................................................... 71 5.3.1 Address Error Sources ............................................................................................ 71 5.3.2 Address Error Exception Source............................................................................. 71 Interrupts.............................................................................................................................. 72 5.4.1 Interrupt Sources..................................................................................................... 72 5.4.2 Interrupt Priority ..................................................................................................... 73 5.4.3 Interrupt Exception Handling ................................................................................. 73 Exceptions Triggered by Instructions .................................................................................. 74 5.5.1 Types of Exceptions Triggered by Instructions ...................................................... 74 5.5.2 Trap Instructions ..................................................................................................... 74 5.5.3 Illegal Slot Instructions ........................................................................................... 75 5.5.4 General Illegal Instructions..................................................................................... 75 Cases when Exceptions are Accepted .................................................................................. 76 Stack States after Exception Handling Ends ........................................................................ 77 Usage Notes ......................................................................................................................... 79 5.8.1 Value of Stack Pointer (SP) .................................................................................... 79 5.8.2 Value of Vector Base Register (VBR) .................................................................... 79 5.8.3 Address Errors Caused by Stacking for Address Error Exception Handling .......... 79 5.8.4 Notes on Slot Illegal Instruction Exception Handling ............................................ 79
Rev. 6.00 Jun. 12, 2007 Page xi of xxxii
5.2
5.3
5.4
5.5
5.6 5.7 5.8
Section 6 Interrupt Controller (INTC)................................................................... 81
6.1 6.2 6.3 Features................................................................................................................................ 81 Input/Output Pins................................................................................................................. 83 Register Descriptions........................................................................................................... 83 6.3.1 Interrupt Control Register 0 (ICR0)........................................................................ 84 6.3.2 IRQ Control Register (IRQCR) .............................................................................. 85 6.3.3 IRQ Status register (IRQSR) .................................................................................. 88 6.3.4 Interrupt Priority Registers A to E (IPRA to IPRE)................................................ 93 Interrupt Sources.................................................................................................................. 95 6.4.1 External Interrupts .................................................................................................. 95 6.4.2 On-Chip Peripheral Module Interrupts ................................................................... 97 6.4.3 User Break Interrupt ............................................................................................... 97 6.4.4 H-UDI Interrupt ...................................................................................................... 97 Interrupt Exception Handling Vector Table......................................................................... 98 Interrupt Operation ............................................................................................................ 100 6.6.1 Interrupt Sequence ................................................................................................ 100 6.6.2 Stack after Interrupt Exception Handling ............................................................. 102 Interrupt Response Time.................................................................................................... 102
6.4
6.5 6.6
6.7
Section 7 Bus State Controller (BSC) ................................................................. 105
7.1 7.2 7.3 Features.............................................................................................................................. 105 Input/Output Pins............................................................................................................... 108 Area Overview................................................................................................................... 109 7.3.1 Area Division........................................................................................................ 109 7.3.2 Shadow Area......................................................................................................... 109 7.3.3 Address Map......................................................................................................... 110 7.3.4 Area 0 Memory Type and Memory Bus Width .................................................... 112 7.3.5 Data Alignment..................................................................................................... 112 Register Descriptions......................................................................................................... 113 7.4.1 Common Control Register (CMNCR) .................................................................. 114 7.4.2 CSn Space Bus Control Register (CSnBCR) (n = 0, 2, 3, 4, 5B, 6B) ................... 115 7.4.3 CSn Space Wait Control Register (CSnWCR) (n = 0, 3, 4, 5B, 6B) .................... 120 7.4.4 SDRAM Control Register (SDCR)....................................................................... 136 7.4.5 Refresh Timer Control/Status Register (RTCSR)................................................. 137 7.4.6 Refresh Timer Counter (RTCNT)......................................................................... 139 7.4.7 Refresh Time Constant Register (RTCOR) .......................................................... 140 Operation ........................................................................................................................... 141 7.5.1 Endian/Access Size and Data Alignment.............................................................. 141 7.5.2 Normal Space Interface ........................................................................................ 146 7.5.3 Access Wait Control ............................................................................................. 150
7.4
7.5
Rev. 6.00 Jun. 12, 2007 Page xii of xxxii
7.5.4 7.5.5 7.5.6 7.5.7 7.5.8 7.5.9
Extension of Chip Select (CSn) Assertion Period................................................. 152 SDRAM Interface ................................................................................................. 153 Byte-Selection SRAM Interface ........................................................................... 179 PCMCIA Interface................................................................................................ 183 Wait between Access Cycles ................................................................................ 190 Others.................................................................................................................... 190
Section 8 Clock Pulse Generator (CPG)..............................................................193
8.1 8.2 8.3 8.4 Features.............................................................................................................................. 193 Input/Output Pins ............................................................................................................... 196 Clock Operating Modes ..................................................................................................... 196 Register Descriptions ......................................................................................................... 198 8.4.1 Frequency Control Register (FRQCR) ................................................................. 198 8.4.2 PHY-LSI Clock Frequency Control Register (MCLKCR) ................................... 200 8.4.3 Usage Notes .......................................................................................................... 201 Changing Frequency .......................................................................................................... 202 8.5.1 Changing Multiplication Ratio ............................................................................. 202 8.5.2 Changing Division Ratio....................................................................................... 203 8.5.3 Changing Clock Operating Mode ......................................................................... 203 Notes on Board Design ...................................................................................................... 205
8.5
8.6
Section 9 Watchdog Timer (WDT) .....................................................................207
9.1 9.2 Features.............................................................................................................................. 207 Register Descriptions ......................................................................................................... 209 9.2.1 Watchdog Timer Counter (WTCNT).................................................................... 209 9.2.2 Watchdog Timer Control/Status Register (WTCSR)............................................ 209 9.2.3 Notes on Register Access...................................................................................... 211 WDT Operation ................................................................................................................. 212 9.3.1 Canceling Software Standbys ............................................................................... 212 9.3.2 Changing Frequency ............................................................................................. 213 9.3.3 Using Watchdog Timer Mode .............................................................................. 213 9.3.4 Using Interval Timer Mode .................................................................................. 214 Usage Note......................................................................................................................... 214
9.3
9.4
Section 10 Power-Down Modes ..........................................................................215
10.1 Features.............................................................................................................................. 215 10.1.1 Types of Power-Down Modes .............................................................................. 215 10.2 Input/Output Pins ............................................................................................................... 217 10.3 Register Descriptions ......................................................................................................... 217 10.3.1 Standby Control Register (STBCR)...................................................................... 218
Rev. 6.00 Jun. 12, 2007 Page xiii of xxxii
10.3.2 Standby Control Register 2 (STBCR2)................................................................. 219 10.3.3 Standby Control Register 3 (STBCR3)................................................................. 220 10.3.4 Standby Control Register 4 (STBCR4)................................................................. 221 10.4 Sleep Mode ........................................................................................................................ 222 10.4.1 Transition to Sleep Mode...................................................................................... 222 10.4.2 Canceling Sleep Mode .......................................................................................... 222 10.5 Software Standby Mode..................................................................................................... 223 10.5.1 Transition to Software Standby Mode .................................................................. 223 10.5.2 Canceling Software Standby Mode ...................................................................... 224 10.6 Module Standby Mode....................................................................................................... 225 10.6.1 Transition to Module Standby Mode .................................................................... 225 10.6.2 Canceling Module Standby Function.................................................................... 225
Section 11 Ethernet Controller (EtherC) ............................................................. 227
11.1 Features.............................................................................................................................. 227 11.2 Input/Output Pins............................................................................................................... 229 11.3 Register Description .......................................................................................................... 231 11.3.1 EtherC Mode Register (ECMR)............................................................................ 232 11.3.2 EtherC Status Register (ECSR) ............................................................................ 235 11.3.3 EtherC Interrupt Permission Register (ECSIPR) .................................................. 237 11.3.4 PHY Interface Register (PIR) ............................................................................... 238 11.3.5 MAC Address High Register (MAHR) ................................................................ 239 11.3.6 MAC Address Low Register (MALR) ................................................................. 239 11.3.7 Receive Frame Length Register (RFLR) .............................................................. 240 11.3.8 PHY Status Register (PSR)................................................................................... 241 11.3.9 Transmit Retry Over Counter Register (TROCR) ................................................ 241 11.3.10 Delayed Collision Detect Counter Register (CDCR)............................................ 242 11.3.11 Lost Carrier Counter Register (LCCR)................................................................. 242 11.3.12 Carrier Not Detect Counter Register (CNDCR) ................................................... 242 11.3.13 CRC Error Frame Counter Register (CEFCR) ..................................................... 243 11.3.14 Frame Receive Error Counter Register (FRECR)................................................. 243 11.3.15 Too-Short Frame Receive Counter Register (TSFRCR) ...................................... 243 11.3.16 Too-Long Frame Receive Counter Register (TLFRCR) ...................................... 244 11.3.17 Residual-Bit Frame Counter Register (RFCR) ..................................................... 244 11.3.18 Multicast Address Frame Counter Register (MAFCR) ........................................ 244 11.3.19 IPG Register (IPGR)............................................................................................. 245 11.3.20 Automatic PAUSE Frame Set Register (APR) ..................................................... 245 11.3.21 Manual PAUSE Frame Set Register (MPR) ......................................................... 246 11.3.22 Automatic PAUSE Frame Retransfer Count Set Register (TPAUSER)............... 246 11.4 Operation ........................................................................................................................... 247
Rev. 6.00 Jun. 12, 2007 Page xiv of xxxii
11.4.1 Transmission......................................................................................................... 247 11.4.2 Reception .............................................................................................................. 249 11.4.3 MII Frame Timing ................................................................................................ 250 11.4.4 Accessing MII Registers ....................................................................................... 252 11.4.5 Magic Packet Detection ........................................................................................ 255 11.4.6 Operation by IPG Setting...................................................................................... 256 11.4.7 Flow Control......................................................................................................... 256 11.5 Connection to PHY-LSI..................................................................................................... 257 11.6 Usage Notes ....................................................................................................................... 258
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC) .......................................................................................259
12.1 Features.............................................................................................................................. 259 12.2 Register Descriptions ......................................................................................................... 260 12.2.1 E-DMAC Mode Register (EDMR) ....................................................................... 261 12.2.2 E-DMAC Transmit Request Register (EDTRR)................................................... 262 12.2.3 E-DMAC Receive Request Register (EDRRR) .................................................... 263 12.2.4 Transmit Descriptor List Address Register (TDLAR).......................................... 264 12.2.5 Receive Descriptor List Address Register (RDLAR) ........................................... 264 12.2.6 EtherC/E-DMAC Status Register (EESR) ............................................................ 265 12.2.7 EtherC/E-DMAC Status Interrupt Permission Register (EESIPR) ....................... 270 12.2.8 Transmit/Receive Status Copy Enable Register (TRSCER)................................. 273 12.2.9 Receive Missed-Frame Counter Register (RMFCR) ............................................ 275 12.2.10 Transmit FIFO Threshold Register (TFTR).......................................................... 275 12.2.11 FIFO Depth Register (FDR) ................................................................................. 277 12.2.12 Receiving method Control Register (RMCR)....................................................... 278 12.2.13 E-DMAC Operation Control Register (EDOCR) ................................................. 279 12.2.14 Receiving-Buffer Write Address Register (RBWAR) .......................................... 280 12.2.15 Receiving-Descriptor Fetch Address Register (RDFAR) ..................................... 280 12.2.16 Transmission-Buffer Read Address Register (TBRAR)....................................... 280 12.2.17 Transmission-Descriptor Fetch Address Register (TDFAR) ................................ 281 12.2.18 Flow Control FIFO Threshold Register (FCFTR) ................................................ 281 12.2.19 Transmit Interrupt Register (TRIMD) .................................................................. 282 12.3 Operation ........................................................................................................................... 283 12.3.1 Descriptor List and Data Buffers .......................................................................... 283 12.3.2 Transmission......................................................................................................... 291 12.3.3 Reception .............................................................................................................. 293 12.3.4 Multi-Buffer Frame Transmit/Receive Processing ............................................... 295 12.4 Usage Notes ....................................................................................................................... 297 12.4.1 Usage Notes on SH-Ether EtherC/E-DMAC Status Register (EESR).................. 297
Rev. 6.00 Jun. 12, 2007 Page xv of xxxii
12.4.2 Usage Notes on SH-Ether Transmit-FIFO Underflow.......................................... 306
Section 13 Compare Match Timer (CMT) .......................................................... 315
13.1 Features.............................................................................................................................. 315 13.2 Register Descriptions......................................................................................................... 316 13.2.1 Compare Match Timer Start Register (CMSTR) .................................................. 316 13.2.2 Compare Match Timer Control/Status Register (CMCSR) .................................. 317 13.2.3 Compare Match Counter (CMCNT)..................................................................... 318 13.2.4 Compare Match Constant Register (CMCOR) ..................................................... 318 13.3 Operation ........................................................................................................................... 319 13.3.1 Interval Count Operation ...................................................................................... 319 13.3.2 CMCNT Count Timing......................................................................................... 319 13.4 Interrupts............................................................................................................................ 320 13.4.1 Interrupt Sources................................................................................................... 320 13.4.2 Timing of Setting Compare Match Flag ............................................................... 320 13.4.3 Timing of Clearing Compare Match Flag............................................................. 320 13.5 Usage Notes ....................................................................................................................... 321 13.5.1 Conflict between Write and Compare-Match Processes of CMCNT ................... 321 13.5.2 Conflict between Word-Write and Count-Up Processes of CMCNT ................... 322 13.5.3 Conflict between Byte-Write and Count-Up Processes of CMCNT..................... 323 13.5.4 Conflict between Write Processes to CMCNT with the Counting Stopped and CMCOR................................................................................................................ 323
Section 14 Serial Communication Interface with FIFO (SCIF).......................... 325
14.1 Overview............................................................................................................................ 325 14.1.1 Features................................................................................................................. 325 14.2 Pin Configuration............................................................................................................... 328 14.3 Register Description .......................................................................................................... 329 14.3.1 Receive Shift Register (SCRSR) .......................................................................... 330 14.3.2 Receive FIFO Data Register (SCFRDR) .............................................................. 330 14.3.3 Transmit Shift Register (SCTSR) ......................................................................... 330 14.3.4 Transmit FIFO Data Register (SCFTDR)............................................................. 331 14.3.5 Serial Mode Register (SCSMR)............................................................................ 331 14.3.6 Serial Control Register (SCSCR).......................................................................... 334 14.3.7 Serial Status Register (SCFSR) ............................................................................ 338 14.3.8 Bit Rate Register (SCBRR) .................................................................................. 346 14.3.9 FIFO Control Register (SCFCR) .......................................................................... 353 14.3.10 FIFO Data Count Register (SCFDR).................................................................... 356 14.3.11 Serial Port Register (SCSPTR) ............................................................................. 357 14.3.12 Line Status Register (SCLSR) .............................................................................. 361
Rev. 6.00 Jun. 12, 2007 Page xvi of xxxii
14.4 Operation ........................................................................................................................... 362 14.4.1 Overview............................................................................................................... 362 14.4.2 Operation in Asynchronous Mode ........................................................................ 364 14.4.3 Synchronous Mode ............................................................................................... 375 14.5 SCIF Interrupts .................................................................................................................. 383 14.6 Serial Port Register (SCSPTR) and SCIF Pins .................................................................. 384 14.7 Usage Notes ....................................................................................................................... 388
Section 15 Host Interface (HIF)...........................................................................391
15.1 Features.............................................................................................................................. 391 15.2 Input/Output Pins ............................................................................................................... 393 15.3 Parallel Access ................................................................................................................... 394 15.3.1 Operation .............................................................................................................. 394 15.3.2 Connection Method............................................................................................... 394 15.4 Register Descriptions ......................................................................................................... 395 15.4.1 HIF Index Register (HIFIDX) .............................................................................. 395 15.4.2 HIF General Status Register (HIFGSR)................................................................ 398 15.4.3 HIF Status/Control Register (HIFSCR) ................................................................ 398 15.4.4 HIF Memory Control Register (HIFMCR) ........................................................... 401 15.4.5 HIF Internal Interrupt Control Register (HIFIICR) .............................................. 403 15.4.6 HIF External Interrupt Control Register (HIFEICR) ............................................ 403 15.4.7 HIF Address Register (HIFADR) ......................................................................... 404 15.4.8 HIF Data Register (HIFDATA) ............................................................................ 405 15.4.9 HIF Boot Control Register (HIFBCR).................................................................. 405 15.4.10 HIFDREQ Trigger Register (HIFDTR) ................................................................ 406 15.4.11 HIF Bank Interrupt Control Register (HIFBICR) ................................................. 407 15.5 Memory Map ..................................................................................................................... 409 15.6 Interface (Basic)................................................................................................................. 410 15.7 Interface (Details) .............................................................................................................. 411 15.7.1 HIFIDX Write/HIFGSR Read .............................................................................. 411 15.7.2 Reading/Writing of HIF Registers other than HIFIDX and HIFGSR................... 411 15.7.3 Consecutive Data Writing to HIFRAM by External Device................................. 412 15.7.4 Consecutive Data Reading from HIFRAM to External Device ............................ 412 15.8 External DMAC Interface.................................................................................................. 413 15.9 Interface When External Device Power is Cut Off ............................................................ 419
Section 16 Pin Function Controller (PFC)...........................................................423
16.1 Register Descriptions ......................................................................................................... 432 16.1.1 Port A IO Register H (PAIORH) .......................................................................... 433 16.1.2 Port A Control Register H1 and H2 (PACRH1 and PACRH2) ............................ 433
Rev. 6.00 Jun. 12, 2007 Page xvii of xxxii
16.1.3 Port B IO Register L (PBIORL) ........................................................................... 436 16.1.4 Port B Control Register L1 and L2 (PBCRL1 and PBCRL2)............................... 436 16.1.5 Port C IO Register H and L (PCIORH and PCIORL) .......................................... 440 16.1.6 Port C Control Register H2, L1, and L2 (PCCRH2, PCCRL1, and PCCRL2)..... 440 16.1.7 Port D IO Register L (PDIORL)........................................................................... 445 16.1.8 Port D Control Register L2 (PDCRL2) ................................................................ 446 16.1.9 Port E IO Register H and L (PEIORH and PEIORL) ........................................... 448 16.1.10 Port E Control Register H1, H2, L1, and L2 (PECRH1, PECRH2, PECRL1, and PECRL2) .............................................................................................................. 448
Section 17 I/O Ports............................................................................................. 457
17.1 Port A................................................................................................................................. 457 17.1.1 Register Description ............................................................................................. 457 17.1.2 Port A Data Register H (PADRH) ........................................................................ 457 17.2 Port B ................................................................................................................................. 459 17.2.1 Register Description ............................................................................................. 459 17.2.2 Port B Data Register L (PBDRL) ......................................................................... 459 17.3 Port C ................................................................................................................................. 461 17.3.1 Register Description ............................................................................................. 462 17.3.2 Port C Data Registers H and L (PCDRH and PCDRL) ........................................ 462 17.4 Port D................................................................................................................................. 464 17.4.1 Register Description ............................................................................................. 464 17.4.2 Port D Data Register L (PDDRL)......................................................................... 464 17.5 Port E ................................................................................................................................. 466 17.5.1 Register Description ............................................................................................. 467 17.5.2 Port E Data Registers H and L (PEDRH and PEDRL) ......................................... 467 17.6 Usage Note......................................................................................................................... 469
Section 18 User Break Controller (UBC)............................................................ 471
18.1 Features.............................................................................................................................. 471 18.2 Register Descriptions......................................................................................................... 473 18.2.1 Break Address Register A (BARA)...................................................................... 473 18.2.2 Break Address Mask Register A (BAMRA)......................................................... 474 18.2.3 Break Bus Cycle Register A (BBRA)................................................................... 474 18.2.4 Break Address Register B (BARB) ...................................................................... 475 18.2.5 Break Address Mask Register B (BAMRB) ......................................................... 476 18.2.6 Break Data Register B (BDRB)............................................................................ 476 18.2.7 Break Data Mask Register B (BDMRB)............................................................... 477 18.2.8 Break Bus Cycle Register B (BBRB) ................................................................... 477 18.2.9 Break Control Register (BRCR) ........................................................................... 479
Rev. 6.00 Jun. 12, 2007 Page xviii of xxxii
18.2.10 Execution Times Break Register (BETR)............................................................. 482 18.2.11 Branch Source Register (BRSR)........................................................................... 482 18.2.12 Branch Destination Register (BRDR)................................................................... 483 18.3 Operation ........................................................................................................................... 484 18.3.1 Flow of User Break Operation .............................................................................. 484 18.3.2 Break on Instruction Fetch Cycle.......................................................................... 485 18.3.3 Break on Data Access Cycle................................................................................. 485 18.3.4 Sequential Break ................................................................................................... 486 18.3.5 Value of Saved Program Counter (PC)................................................................. 486 18.3.6 PC Trace ............................................................................................................... 487 18.3.7 Usage Examples.................................................................................................... 488 18.3.8 Usage Notes .......................................................................................................... 492
Section 19 User Debugging Interface (H-UDI) ...................................................493
19.1 Features.............................................................................................................................. 493 19.2 Input/Output Pins ............................................................................................................... 495 19.3 Register Descriptions ......................................................................................................... 496 19.3.1 Bypass Register (SDBPR) .................................................................................... 496 19.3.2 Instruction Register (SDIR) .................................................................................. 496 19.3.3 Boundary Scan Register (SDBSR) ....................................................................... 497 19.3.4 ID Register (SDID)............................................................................................... 503 19.4 Operation ........................................................................................................................... 504 19.4.1 TAP Controller ..................................................................................................... 504 19.4.2 Reset Configuration .............................................................................................. 505 19.4.3 TDO Output Timing ............................................................................................. 505 19.4.4 H-UDI Reset ......................................................................................................... 506 19.4.5 H-UDI Interrupt .................................................................................................... 506 19.5 Boundary Scan ................................................................................................................... 507 19.5.1 Supported Instructions .......................................................................................... 507 19.5.2 Points for Attention............................................................................................... 508 19.6 Usage Notes ....................................................................................................................... 508
Section 20 List of Registers .................................................................................509
20.1 Register Addresses (Address Order).................................................................................. 510 20.2 Register Bits....................................................................................................................... 516 20.3 Register States in Each Processing State ........................................................................... 533
Section 21 Electrical Characteristics ...................................................................539
21.1 Absolute Maximum Ratings .............................................................................................. 539 21.2 Power-On and Power-Off Order ........................................................................................ 540
Rev. 6.00 Jun. 12, 2007 Page xix of xxxii
21.3 DC Characteristics ............................................................................................................. 542 21.4 AC Characteristics ............................................................................................................. 544 21.4.1 Clock Timing ........................................................................................................ 545 21.4.2 Control Signal Timing .......................................................................................... 549 21.4.3 Bus Timing ........................................................................................................... 551 21.4.4 Basic Timing......................................................................................................... 553 21.4.5 Synchronous DRAM Timing................................................................................ 559 21.4.6 PCMCIA Timing .................................................................................................. 576 21.4.7 SCIF Timing ......................................................................................................... 580 21.4.8 Port Timing........................................................................................................... 581 21.4.9 HIF Timing ........................................................................................................... 582 21.4.10 EtherC Timing ...................................................................................................... 585 21.4.11 H-UDI Related Pin Timing................................................................................... 588 21.4.12 AC Characteristic Test Conditions ....................................................................... 590 21.4.13 Delay Time Variation Due to Load Capacitance (Reference Values) .................. 591
Appendix
A. B. C.
......................................................................................................... 593
Port States in Each Pin State.............................................................................................. 593 Product Code Lineup ......................................................................................................... 597 Package Dimensions .......................................................................................................... 598
Main Revisions and Additions in this Edition..................................................... 599 Index ......................................................................................................... 605
Rev. 6.00 Jun. 12, 2007 Page xx of xxxii
Figures
Section 1 Overview Figure 1.1 Block Diagram .............................................................................................................. 6 Figure 1.2 Pin Assignments ............................................................................................................ 7 Section 2 Figure 2.1 Figure 2.2 Figure 2.3 Figure 2.4 Section 3 Figure 3.1 Figure 3.2 Figure 3.3 Figure 3.4 Section 6 Figure 6.1 Figure 6.2 Figure 6.3 Figure 6.4 Section 7 Figure 7.1 Figure 7.2 Figure 7.3 Figure 7.4 CPU CPU Internal Register Configuration .......................................................................... 20 Register Data Format................................................................................................... 24 Memory Data Format .................................................................................................. 24 CPU State Transition................................................................................................... 47 Cache Cache Structure ........................................................................................................... 49 Cache Search Scheme ................................................................................................. 55 Write-Back Buffer Configuration................................................................................ 57 Specifying Address and Data for Memory-Mapped Cache Access............................. 60 Interrupt Controller (INTC) INTC Block Diagram .................................................................................................. 82 Block Diagram of IRQ7 to IRQ0 Interrupts Control................................................... 96 Interrupt Sequence Flowchart.................................................................................... 101 Stack after Interrupt Exception Handling .................................................................. 102
Bus State Controller (BSC) Block Diagram of BSC.............................................................................................. 107 Address Space ........................................................................................................... 110 Normal Space Basic Access Timing (No-Wait Access)............................................ 146 Consecutive Access to Normal Space (1): Bus Width = 16 bits, Longword Access, CSnWCR.WM = 0 (Access Wait = 0, Cycle Wait = 0) ............. 147 Figure 7.5 Consecutive Access to Normal Space (2): Bus Width = 16 bits, Longword Access, CSnWCR.WM = 1 (Access Wait = 0, Cycle Wait = 0) ............. 148 Figure 7.6 Example of 16-Bit Data-Width SRAM Connection .................................................. 149 Figure 7.7 Example of 8-Bit Data-Width SRAM Connection.................................................... 149 Figure 7.8 Wait Timing for Normal Space Access (Software Wait Only) ................................. 150 Figure 7.9 Wait Cycle Timing for Normal Space Access (Wait cycle Insertion using WAIT).. 151 Figure 7.10 Example of Timing when CSn Assertion Period is Extended ................................. 152 Figure 7.11 Example of 16-Bit Data-Width SDRAM Connection ............................................. 154 Figure 7.12 Burst Read Basic Timing (Auto Precharge) ............................................................ 162 Figure 7.13 Burst Read Wait Specification Timing (Auto Precharge) ....................................... 163 Figure 7.14 Basic Timing for Single Read (Auto Precharge) ..................................................... 164
Rev. 6.00 Jun. 12, 2007 Page xxi of xxxii
Figure 7.15 Figure 7.16 Figure 7.17 Figure 7.18 Figure 7.19 Figure 7.20 Figure 7.21 Figure 7.22 Figure 7.23 Figure 7.24 Figure 7.25 Figure 7.26 Figure 7.27 Figure 7.28 Figure 7.29 Figure 7.30 Figure 7.31 Figure 7.32 Figure 7.33 Figure 7.34 Figure 7.35 Figure 7.36
Basic Timing for Burst Write (Auto Precharge) ..................................................... 166 Basic Timing for Single Write (Auto-Precharge).................................................... 167 Burst Read Timing (No Auto Precharge) ................................................................ 169 Burst Read Timing (Bank Active, Same Row Address) ......................................... 170 Burst Read Timing (Bank Active, Different Row Addresses) ................................ 171 Single Write Timing (No Auto Precharge).............................................................. 172 Single Write Timing (Bank Active, Same Row Address)....................................... 173 Single Write Timing (Bank Active, Different Row Addresses).............................. 174 Auto-Refreshing Timing ......................................................................................... 175 Self-Refreshing Timing........................................................................................... 177 Write Timing for SDRAM Mode Register (Based on JEDEC)............................... 179 Basic Access Timing for Byte-Selection SRAM (BAS = 0)................................... 180 Basic Access Timing for Byte-Selection SRAM (BAS = 1)................................... 181 Wait Timing for Byte-Selection SRAM (BAS = 1) (Software Wait Only)............. 182 Example of Connection with 16-Bit Data-Width Byte-Selection SRAM ............... 183 Example of PCMCIA Interface Connection............................................................ 184 Basic Access Timing for PCMCIA Memory Card Interface................................... 185 Wait Timing for PCMCIA Memory Card Interface (TED[3:0] = B'0010, TEH[3:0] = B'0001, Software Wait = 1, Hardware Wait = 1)................................. 186 Example of PCMCIA Space Assignment (CS5BWCR.SA[1:0] = B'10, CS6BWCR.SA[1:0] = B'10) ................................................................................... 187 Basic Timing for PCMCIA I/O Card Interface ....................................................... 188 Wait Timing for PCMCIA I/O Card Interface (TED[3:0] = B'0010, TEH[3:0] = B'0001, Software Wait = 1, Hardware Wait = 1) ................................ 189 Timing for Dynamic Bus Sizing of PCMCIA I/O Card Interface (TED[3:0] = B'0010, TEH[3:0] = B'0001, Software Waits = 3)............................. 189
Section 8 Clock Pulse Generator (CPG) Figure 8.1 Block Diagram of CPG ............................................................................................. 194 Figure 8.2 Points for Attention when Using Crystal Resonator.................................................. 205 Section 9 Watchdog Timer (WDT) Figure 9.1 Block Diagram of WDT ............................................................................................ 208 Figure 9.2 Writing to WTCNT and WTCSR.............................................................................. 212 Section 10 Power-Down Modes Figure 10.1 Canceling Standby Mode with STBY Bit in STBCR.............................................. 225 Section 11 Figure 11.1 Figure 11.2 Figure 11.3 Ethernet Controller (EtherC) Configuration of EtherC.......................................................................................... 228 EtherC Transmitter State Transitions ...................................................................... 248 EtherC Receiver State Transmissions ..................................................................... 249
Rev. 6.00 Jun. 12, 2007 Page xxii of xxxii
Figure 11.4 (1) MII Frame Transmit Timing (Normal Transmission)........................................ 250 Figure 11.4 (2) MII Frame Transmit Timing (Collision)............................................................ 250 Figure 11.4 (3) MII Frame Transmit Timing (Transmit Error)................................................... 251 Figure 11.4 (4) MII Frame Receive Timing (Normal Reception)............................................... 251 Figure 11.4 (5) MII Frame Receive Timing (Reception Error (1))............................................. 251 Figure 11.4 (6) MII Fame Receive Timing (Reception Error (2)) .............................................. 251 Figure 11.5 MII Management Frame Format ............................................................................. 252 Figure 11.6 (1) 1-Bit Data Write Flowchart ............................................................................... 253 Figure 11.6 (2) Bus Release Flowchart (TA in Read in Figure 11.5) ......................................... 254 Figure 11.6 (3) 1-Bit Data Read Flowchart ................................................................................ 254 Figure 11.6 (4) Independent Bus Release Flowchart (IDLE in Write in Figure 11.5)................ 255 Figure 11.7 Changing IPG and Transmission Efficiency ........................................................... 256 Figure 11.8 Example of Connection to DP83846AVHG............................................................ 257 Section 12 Figure 12.1 Figure 12.2 Figure 12.3 Figure 12.4 Figure 12.5 Figure 12.6 Figure 12.7 Figure 12.8 Ethernet Controller Direct Memory Access Controller (E-DMAC) Configuration of E-DMAC, and Descriptors and Buffers....................................... 259 Relationship between Transmit Descriptor and Transmit Buffer ............................ 283 Relationship between Receive Descriptor and Receive Buffer ............................... 287 Sample Transmission Flowchart ............................................................................. 292 Sample Reception Flowchart................................................................................... 294 E-DMAC Operation after Transmit Error ............................................................... 295 E-DMAC Operation after Receive Error................................................................. 296 Timing of the Case where Setting of the Interrupt Source Bit in EESR by the E-DMAC Fails............................................................................................. 297 Figure 12.9 Countermeasure by Monitoring the Transmit Descriptor in Processing of Interrupts Other than the Frame Transmit Complete (TC) Interrupt.................. 303 Figure 12.10 Method of Adding Timeout Processing................................................................. 305 Figure 12.11 Operation when E-DMAC Stops and the Transmit FIFO ..................................... 307 Figure 12.12 Processing Transmission without Handling of the TC Interrupt ........................... 310 Figure 12.13 Countermeasure for the Case with TC Interrupt-Driven Software: Addition of Timeout Processing within the Limit Imposed by the Maximum Specified Time..................................................................................... 313 Section 13 Figure 13.1 Figure 13.2 Figure 13.3 Figure 13.4 Figure 13.5 Figure 13.6 Figure 13.7 Compare Match Timer (CMT) Block Diagram of Compare Match Timer............................................................... 315 Counter Operation ................................................................................................... 319 Count Timing .......................................................................................................... 319 Timing of CMF Setting ........................................................................................... 320 Conflict between Write and Compare-Match Processes of CMCNT ...................... 321 Conflict between Word-Write and Count-Up Processes of CMCNT...................... 322 Conflict between Byte-Write and Count-Up Processes of CMCNT ....................... 323
Rev. 6.00 Jun. 12, 2007 Page xxiii of xxxii
Section 14 Serial Communication Interface with FIFO (SCIF) Figure 14.1 Block Diagram of SCIF........................................................................................... 327 Figure 14.2 Example of Data Format in Asynchronous Communication (8-Bit Data with Parity and Two Stop Bits)............................................................ 364 Figure 14.3 Sample Flowchart for SCIF Initialization ............................................................... 367 Figure 14.4 Sample Flowchart for Transmitting Serial Data...................................................... 368 Figure 14.5 Example of Transmit Operation (8-Bit Data, Parity, One Stop Bit)........................ 370 Figure 14.6 Example of Operation Using Modem Control (CTS).............................................. 370 Figure 14.7 Sample Flowchart for Receiving Serial Data .......................................................... 371 Figure 14.8 Sample Flowchart for Receiving Serial Data (cont)................................................ 372 Figure 14.9 Example of SCIF Receive Operation (8-Bit Data, Parity, One Stop Bit)................ 374 Figure 14.10 Example of Operation Using Modem Control (RTS)............................................ 374 Figure 14.11 Data Format in Synchronous Communication ...................................................... 375 Figure 14.12 Sample Flowchart for SCIF Initialization ............................................................. 377 Figure 14.13 Sample Flowchart for Transmitting Serial Data.................................................... 378 Figure 14.14 Example of SCIF Transmit Operation................................................................... 379 Figure 14.15 Sample Flowchart for Receiving Serial Data (1)................................................... 380 Figure 14.16 Sample Flowchart for Receiving Serial Data (2)................................................... 380 Figure 14.17 Example of SCIF Receive Operation .................................................................... 381 Figure 14.18 Sample Flowchart for Transmitting/Receiving Serial Data................................... 382 Figure 14.19 RTSIO Bit, RTSDT bit, and RTS Pin ................................................................... 384 Figure 14.20 CTSIO Bit, CTSDT bit, and CTS Pin ................................................................... 385 Figure 14.21 SCKIO Bit, SCKDT bit, and SCK Pin .................................................................. 386 Figure 14.22 SPBIO Bit, SPBDT bit, and TxD Pin.................................................................... 386 Figure 14.23 SPBDT bit and RxD Pin........................................................................................ 387 Figure 14.24 Receive Data Sampling Timing in Asynchronous Mode ...................................... 389 Section 15 Host Interface (HIF) Figure 15.1 Block Diagram of HIF............................................................................................. 392 Figure 15.2 HIF Connection Example........................................................................................ 394 Figure 15.3 Basic Timing for HIF Interface ............................................................................... 410 Figure 15.4 HIFIDX Write and HIFGSR Read .......................................................................... 411 Figure 15.5 HIF Register Settings .............................................................................................. 411 Figure 15.6 Consecutive Data Writing to HIFRAM................................................................... 412 Figure 15.7 Consecutive Data Reading from HIFRAM ............................................................. 413 Figure 15.8 HIFDREQ Timing (When DMD = 0 and DPOL = 0)............................................. 413 Figure 15.9 HIFDREQ Timing (When DMD = 0 and DPOL = 1)............................................. 414 Figure 15.10 HIFDREQ Timing (When DMD = 1 and DPOL = 0) ........................................... 414 Figure 15.11 HIFDREQ Timing (When DMD = 1 and DPOL = 1) ........................................... 415 Figure 15.12 Image of High-Impedance Control of HIF Pins by HIFEBL Pin .......................... 419
Rev. 6.00 Jun. 12, 2007 Page xxiv of xxxii
Section 17 Figure 17.1 Figure 17.2 Figure 17.3 Figure 17.4 Figure 17.5
I/O Ports Port A ...................................................................................................................... 457 Port B ...................................................................................................................... 459 Port C ...................................................................................................................... 461 Port D ...................................................................................................................... 464 Port E....................................................................................................................... 466
Section 18 User Break Controller (UBC) Figure 18.1 Block Diagram of UBC........................................................................................... 472 Section 19 Figure 19.1 Figure 19.2 Figure 19.3 Figure 19.4 User Debugging Interface (H-UDI) Block Diagram of H-UDI........................................................................................ 494 TAP Controller State Transitions ............................................................................ 504 H-UDI Data Transfer Timing.................................................................................. 506 H-UDI Reset............................................................................................................ 506
Section 21 Electrical Characteristics Figure 21.1 External Clock Input Timing................................................................................... 546 Figure 21.2 CKIO and CK_PHY Clock Output Timings ........................................................... 546 Figure 21.3 Oscillation Settling Timing after Power-On............................................................ 547 Figure 21.4 Oscillation Settling Timing after Standby Mode (By Reset)................................... 547 Figure 21.5 Oscillation Settling Timing after Standby Mode (By NMI or IRQ)........................ 547 Figure 21.6 PLL Synchronize Settling Timing By Reset or NMI .............................................. 548 Figure 21.7 Reset Input Timing.................................................................................................. 549 Figure 21.8 Interrupt Input Timing............................................................................................. 550 Figure 21.9 Pin Drive Timing in Standby Mode ........................................................................ 550 Figure 21.10 Basic Bus Timing: No Wait Cycle ........................................................................ 553 Figure 21.11 Basic Bus Timing: One Software Wait Cycle ....................................................... 554 Figure 21.12 Basic Bus Timing: One External Wait Cycle ........................................................ 555 Figure 21.13 Basic Bus Timing: One Software Wait Cycle, External Wait Enabled (WM Bit = 0), No Idle Cycle ................................................................................ 556 Figure 21.14 Byte Control SRAM Timing: SW = 1 Cycle, HW = 1 Cycle, One Asynchronous External Wait Cycle, CSnWCR.BAS = 0 (UB-/LB-Controlled Write Cycle) ........................................................................ 557 Figure 21.15 Byte Control SRAM Timing: SW = 1 Cycle, HW = 1 Cycle, One Asynchronous External Wait Cycle, CSnWCR.BAS = 1 (WE-Controlled Write Cycle) ............................................................................... 558 Figure 21.16 Synchronous DRAM Single Read Bus Cycle (Auto-Precharge, CAS Latency = 2, WTRCD = 0 Cycle, WTRP = 0 Cycle)...... 559 Figure 21.17 Synchronous DRAM Single Read Bus Cycle (Auto-Precharge, CAS Latency = 2, WTRCD = 1 Cycle, WTRP = 1 Cycle)...... 560
Rev. 6.00 Jun. 12, 2007 Page xxv of xxxii
Figure 21.18 Synchronous DRAM Burst Read Bus Cycle (Single Read x 4) (Auto-Precharge, CAS Latency = 2, WTRCD = 0 Cycle, WTRP = 1 Cycle)....... 561 Figure 21.19 Synchronous DRAM Burst Read Bus Cycle (Single Read x 4) (Auto-Precharge, CAS Latency = 2, WTRCD = 1 Cycle, WTRP = 0 Cycle)....... 562 Figure 21.20 Synchronous DRAM Single Write Bus Cycle (Auto-Precharge, TRWL = 1 Cycle)..................................................................... 563 Figure 21.21 Synchronous DRAM Single Write Bus Cycle (Auto-Precharge, WTRCD = 2 Cycles, TRWL = 1 Cycle)................................... 564 Figure 21.22 Synchronous DRAM Burst Write Bus Cycle (Single Write x 4) (Auto-Precharge, WTRCD = 0 Cycle, TRWL = 1 Cycle) .................................... 565 Figure 21.23 Synchronous DRAM Burst Write Bus Cycle (Single Write x 4) (Auto-Precharge, WTRCD = 1 Cycle, TRWL = 1 Cycle) .................................... 566 Figure 21.24 Synchronous DRAM Burst Read Bus Cycle (Single Read x 4) (Bank Active Mode: ACT + READ Commands, CAS Latency = 2, WTRCD = 0 Cycle) .............................................................................................. 567 Figure 21.25 Synchronous DRAM Burst Read Bus Cycle (Single Read x 4) (Bank Active Mode: READ Command, Same Row Address, CAS Latency = 2, WTRCD = 0 Cycle) .............................................................................................. 568 Figure 21.26 Synchronous DRAM Burst Read Bus Cycle (Single Read x 4) (Bank Active Mode: PRE + ACT + READ Commands, Different Row Addresses, CAS Latency = 2, WTRCD = 0 Cycle)................................................................. 569 Figure 21.27 Synchronous DRAM Burst Write Bus Cycle (Single Write x 4) (Bank Active Mode: ACT + WRITE Commands, WTRCD = 0 Cycle, TRWL = 0 Cycle) ................................................................................................. 570 Figure 21.28 Synchronous DRAM Burst Write Bus Cycle (Single Write x 4) (Bank Active Mode: WRITE Command, Same Row Address, WTRCD = 0 Cycle, TRWL = 0 Cycle) ................................................................................................. 571 Figure 21.29 Synchronous DRAM Burst Write Bus Cycle (Single Write x 4) (Bank Active Mode: PRE + ACT + WRITE Commands, Different Row Addresses, WTRCD = 0 Cycle, TRWL = 0 Cycle)................................................................. 572 Figure 21.30 Synchronous DRAM Auto-Refreshing Timing (WTRP = 1 Cycle, WTRC = 3 Cycles)................................................................. 573 Figure 21.31 Synchronous DRAM Self-Refreshing Timing (WTRP = 1 Cycle) ....................... 574 Figure 21.32 Synchronous DRAM Mode Register Write Timing (WTRP = 1 Cycle)............... 575 Figure 21.33 PCMCIA Memory Card Interface Bus Timing ..................................................... 576 Figure 21.34 PCMCIA Memory Card Interface Bus Timing (TED = 2.5 Cycles, TEH = 1.5 Cycles, One Software Wait Cycle, One External Wait Cycle) ........... 577 Figure 21.35 PCMCIA I/O Card Interface Bus Timing.............................................................. 578 Figure 21.36 PCMCIA I/O Card Interface Bus Timing (TED = 2.5 Cycles, TEH = 1.5 Cycles, One Software Wait Cycle, One External Wait Cycle) .......................................... 579
Rev. 6.00 Jun. 12, 2007 Page xxvi of xxxii
Figure 21.37 Figure 21.38 Figure 21.39 Figure 21.40 Figure 21.41 Figure 21.42 Figure 21.43 Figure 21.44 Figure 21.45 Figure 21.46 Figure 21.47 Figure 21.48 Figure 21.49 Figure 21.50 Figure 21.51 Figure 21.52 Figure 21.53 Figure 21.54 Figure 21.55
SCK Input Clock Timing....................................................................................... 580 SCI Input/Output Timing in Clocked Synchronous Mode .................................... 581 I/O Port Timing ..................................................................................................... 581 HIF Access Timing ............................................................................................... 583 HIFINT and HIFDREQ Timing ............................................................................ 583 HIFRDY and HIF Pin Enable/Disable Timing...................................................... 584 MII Transmission Timing (Normal Operation)..................................................... 586 MII Transmission Timing (Collision Occurred).................................................... 586 MII Reception Timing (Normal Operation) .......................................................... 587 MII Reception Timing (Error Occurred) ............................................................... 587 MDIO Input Timing .............................................................................................. 587 MDIO Output Timing ........................................................................................... 587 WOL Output Timing ............................................................................................. 588 EXOUT Output Timing......................................................................................... 588 TCK Input Timing................................................................................................. 589 TCK Input Timing in Reset Hold State ................................................................. 589 H-UDI Data Transmission Timing ........................................................................ 589 Output Load Circuit............................................................................................... 590 Load Capacitance versus Delay Time ................................................................... 591
Appendix Figure C.1 Package Dimensions (BP-176) ................................................................................. 598
Rev. 6.00 Jun. 12, 2007 Page xxvii of xxxii
Rev. 6.00 Jun. 12, 2007 Page xxviii of xxxii
Tables
Section 1 Overview Table 1.1 Pin Functions ............................................................................................................ 8 Table 1.2 Pin Features ............................................................................................................ 13 Section 2 CPU Table 2.1 Initial Values of Registers....................................................................................... 23 Table 2.2 Word Data Sign Extension...................................................................................... 25 Table 2.3 Delayed Branch Instructions................................................................................... 26 Table 2.4 T Bit ........................................................................................................................ 26 Table 2.5 Access to Immediate Data ...................................................................................... 27 Table 2.6 Access to Absolute Address.................................................................................... 27 Table 2.7 Access with Displacement ...................................................................................... 28 Table 2.8 Addressing Modes and Effective Addresses........................................................... 28 Table 2.9 Instruction Formats ................................................................................................. 32 Table 2.10 Instruction Types .................................................................................................... 35 Section 3 Cache Table 3.1 LRU and Way to be Replaced ................................................................................ 50 Table 3.2 Correspondence between Divided Areas and Cache............................................... 51 Section 5 Exception Handling Table 5.1 Types of Exceptions and Priority............................................................................ 65 Table 5.2 Timing for Exception Detection and Start of Exception Handling ......................... 66 Table 5.3 Vector Numbers and Vector Table Address Offsets............................................... 67 Table 5.4 Calculating Exception Handling Vector Table Addresses ...................................... 68 Table 5.5 Reset Status............................................................................................................. 69 Table 5.6 Bus Cycles and Address Errors............................................................................... 71 Table 5.7 Interrupt Sources..................................................................................................... 72 Table 5.8 Interrupt Priority ..................................................................................................... 73 Table 5.9 Types of Exceptions Triggered by Instructions ...................................................... 74 Table 5.10 Delay Slot Instructions, Interrupt Disabled Instructions, and Exceptions............... 76 Table 5.11 Stack Status after Exception Handling Ends........................................................... 77 Section 6 Interrupt Controller (INTC) Table 6.1 Pin Configuration.................................................................................................... 83 Table 6.2 Interrupt Exception Handling Vectors and Priorities.............................................. 98 Table 6.3 Interrupt Response Time....................................................................................... 103 Section 7 Bus State Controller (BSC) Table 7.1 Pin Configuration.................................................................................................. 108
Rev. 6.00 Jun. 12, 2007 Page xxix of xxxii
Table 7.2 Table 7.3 Table 7.4 Table 7.5 Table 7.6 Table 7.7 Table 7.8 Table 7.9 Table 7.10 Table 7.11 Table 7.12 Table 7.13 Table 7.14 Table 7.15 Table 7.16 Table 7.17
Address Map 1 (CMNCR.MAP = 0) .................................................................... 110 Address Map 2 (CMNCR.MAP = 1) .................................................................... 111 Correspondence between External Pin (MD3), Memory Type, and Bus Width for CS0......................................................................................... 112 Correspondence between External Pin (MD5) and Endians................................. 112 16-Bit External Device/Big Endian Access and Data Alignment......................... 142 8-Bit External Device/Big Endian Access and Data Alignment........................... 143 16-Bit External Device/Little Endian Access and Data Alignment ...................... 144 8-Bit External Device/Little Endian Access and Data Alignment ........................ 145 Relationship between Register Settings and Address Multiplex Output (1)......... 155 Relationship between Register Settings and Address Multiplex Output (2)......... 156 Relationship between Register Settings and Address Multiplex Output (3)......... 157 Relationship between Register Settings and Address Multiplex Output (4)......... 158 Relationship between Register Settings and Address Multiplex Output (5)......... 159 Relationship between Register Settings and Address Multiplex Output (6)......... 160 Relationship between Access Size and Number of Bursts.................................... 161 Access Address for SDRAM Mode Register Write.............................................. 178
Section 8 Clock Pulse Generator (CPG) Table 8.1 Pin Configuration.................................................................................................. 196 Table 8.2 Mode Control Pins and Clock Operating Modes .................................................. 196 Table 8.3 Possible Combination of Clock Modes and FRQCR Values................................ 197 Section 10 Power-Down Modes Table 10.1 States of Power-Down Modes .............................................................................. 216 Table 10.2 Pin Configuration.................................................................................................. 217 Table 10.3 Register States in Software Standby Mode........................................................... 223 Section 11 Ethernet Controller (EtherC) Table 11.1 Pin Configuration.................................................................................................. 229 Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC) Table 12.1 EESR Bits for which This Problem can Occur and Reflection of Interrupt Sources in the Descriptor ....................................................................... 298 Table 12.2 Reference Values for Maximum Specified Time.................................................. 314 Section 14 Serial Communication Interface with FIFO (SCIF) Table 14.1 SCIF Pins.............................................................................................................. 328 Table 14.2 SCSMR Settings ................................................................................................... 347 Table 14.3 Bit Rates and SCBRR Settings in Asynchronous Mode....................................... 347 Table 14.4 Bit Rates and SCBRR Settings in Synchronous Mode ......................................... 350 Table 14.5 Maximum Bit Rates for Various Frequencies with Baud Rate Generator (Asynchronous Mode) .......................................................................................... 351
Rev. 6.00 Jun. 12, 2007 Page xxx of xxxii
Table 14.6 Table 14.7 Table 14.8 Table 14.9 Table 14.10 Table 14.11
Maximum Bit Rates with External Clock Input (Asynchronous Mode)............... 352 Maximum Bit Rates with External Clock Input (Synchronous Mode) ................. 352 SCSMR Settings and SCIF Communication Formats........................................... 363 SCSMR and SCSCR Settings and SCIF Clock Source Selection......................... 363 Serial Communication Formats (Asynchronous Mode).................................... 365 SCIF Interrupt Sources ..................................................................................... 383
Section 15 Host Interface (HIF) Table 15.1 Pin Configuration.................................................................................................. 393 Table 15.2 HIF Operations ..................................................................................................... 394 Table 15.3 Memory Map ........................................................................................................ 409 Table 15.4 Consecutive Write Procedure to HIFRAM by External DMAC........................... 416 Table 15.5 Consecutive Read Procedure from HIFRAM by External DMAC....................... 417 Table 15.6 Input/Output Control for HIF Pins........................................................................ 420 Section 16 Pin Function Controller (PFC) Table 16.1 List of Multiplexed Pins (Port A) ......................................................................... 423 Table 16.2 List of Multiplexed Pins (Port B).......................................................................... 423 Table 16.3 List of Multiplexed Pins (Port C).......................................................................... 425 Table 16.4 List of Multiplexed Pins (Port D) ......................................................................... 425 Table 16.5 List of Multiplexed Pins (Port E).......................................................................... 426 Table 16.6 Pin Functions in Each Operating Mode ................................................................ 427 Section 17 I/O Ports Table 17.1 Port A Data Register H (PADRH) Read/Write Operation .................................... 458 Table 17.2 Port B Data Register L (PBDRL) Read/Write Operation ..................................... 460 Table 17.3 Port C Data Registers H and L (PCDRH and PCDRL) Read/Write Operation .... 463 Table 17.4 Port D Data Register L (PDDRL) Read/Write Operation..................................... 465 Table 17.5 Port E Data Registers H, L (PEDRH, PEDRL) Read/Write Operation ................ 468 Section 18 User Break Controller (UBC) Table 18.1 Data Access Cycle Addresses and Operand Size Comparison Conditions ........... 485 Section 19 User Debugging Interface (H-UDI) Table 19.1 Pin Configuration.................................................................................................. 495 Table 19.2 H-UDI Commands................................................................................................ 497 Table 19.3 External pins and Boundary Scan Register Bits ................................................... 498 Table 19.4 Reset Configuration .............................................................................................. 505 Section 21 Electrical Characteristics Table 21.1 Absolute Maximum Ratings ................................................................................. 539 Table 21.2 Recommended Timing at Power-On..................................................................... 540 Table 21.3 Recommended Timing in Power-Off.................................................................... 541
Rev. 6.00 Jun. 12, 2007 Page xxxi of xxxii
Table 21.4 Table 21.4 Table 21.5 Table 21.6 Table 21.7 Table 21.8 Table 21.9 Table 21.10 Table 21.11 Table 21.12 Table 21.13 Table 21.14 Appendix Table A.1
DC Characteristics (1) .......................................................................................... 542 DC Characteristics (2) .......................................................................................... 543 Permissible Output Currents................................................................................. 544 Maximum Operating Frequency ........................................................................... 544 Clock Timing ........................................................................................................ 545 Control Signal Timing .......................................................................................... 549 Bus Timing ........................................................................................................... 551 SCIF Timing ..................................................................................................... 580 Port Timing....................................................................................................... 581 HIF Timing ....................................................................................................... 582 EtherC Timing .................................................................................................. 585 H-UDI Related Pin Timing............................................................................... 588 Port States in Each Pin State................................................................................. 593
Rev. 6.00 Jun. 12, 2007 Page xxxii of xxxii
Section 1 Overview
Section 1 Overview
This LSI is a CMOS single-chip microcontroller that integrates a high-speed CPU core using an original Renesas Technology RISC (Reduced Instruction Set Computer) architecture with supporting functions required for an Ethernet system. The CPU of this LSI has a RISC (Reduced Instruction Set Computer) type instruction set. The CPU basically operates at a rate of one instruction per cycle, offering a great improvement in instruction execution speed. In addition, the 32-bit internal architecture provides improved data processing power. With this CPU, it has become possible to assemble low-cost, highperformance/high-functionality systems even for applications such as realtime control, which could not previously be handled by microcontrollers because of their high-speed processing requirements. This LSI is equipped with a media access controller (MAC) conforming to the IEEE802.3u standard, and an Ethernet controller that includes a media independent interface (MII) standard unit, enabling 10/100 Mbps LAN connection. Supporting functions necessary for system configuration are also provided, including cache memory, RAM, timers, a serial communication interface with FIFO (SCIF), host interface (HFI), interrupt controller (INTC), and I/O ports. The external memory access support function of this LSI enables direct connection to various types of memory, such as standard memory, SDRAM, and PCMCIA. This greatly reduces system cost.
Rev. 6.00 Jun. 12, 2007 Page 1 of 610 REJ09B0131-0600
Section 1 Overview
1.1
Features
The features of this LSI are shown below. CPU: * Central processing unit with an internal 32-bit RISC (Reduced Instruction Set Computer) architecture * Instruction length: 16-bit fixed length for improved code efficiency * Load-store architecture (basic operations are executed between registers) * Sixteen 32-bit general registers * Five-stage pipeline * On-chip multiplier: Multiplication operations (32 bits x 32 bits 64 bits) executed in two to five cycles * C language-oriented 62 basic instructions Note: Some specifications on the slot illegal instruction differ from the conventional SH2 core. For details, see section 5.8, Usage Notes, in section 5, Exception Handling. User break controller (UBC): * Address, data value, access type, and data size are available for setting as break conditions * Supports the sequential break function * Two break channels U memory: * 4 kbytes Cache memory: * * * * Unified cache, mixture of instructions and data 4-way set associative type Selection of write-back or write-through mode 4 kbytes (SH7618), 16kbytes (SH7618A)
Rev. 6.00 Jun. 12, 2007 Page 2 of 610 REJ09B0131-0600
Section 1 Overview
Bus state controller (BSC): * Address space is divided into five areas: three areas 0, 3, and 4; each a maximum of 64 Mbytes, and two areas 5B and 6B; each a maximum of 32 Mbytes (address map 1 mode). * Address space is divided into five areas, 0, 3, 4, 5, and 6; each a maximum of 64 Mbytes (address map 2 mode). * 16-bit external bus * The following features are settable for each area. Bus size (8 or 16 bits) Number of access wait cycles Setting of idle wait cycles Specifying the memory to be connected to each area enables direct connection to SRAM, SDRAM, and PCMCIA. Outputs chip select signals (CS0, CS3, CS4, CS5B, and CS6B) for corresponding area * SDRAM refresh function Supports auto-refresh and self-refresh modes * SDRAM burst access function * PCMCIA access function Conforms to the JEIDA Ver. 4.2 standard, two slots * Selection of big or little endian mode (The mode of all the areas is switched collectively by a mode pin.) Interrupt controller (INTC): * Supports nine external interrupt pins (NMI, IRQ7 to IRQ0) * On-chip peripheral interrupt: Priority level is independently selected for each module * Vector address: Specified vector address for each interrupt source User debugging interface (H-UDI): * Supports the JTAG interface emulator * JTAG standard pins arranged Clock pulse generator (CPG): * Clock mode: Clock source selectable between an external supply and crystal resonator * Three types of clocks generated: CPU clock: 100 MHz (max.) Bus clock: 50 MHz (max.)
Rev. 6.00 Jun. 12, 2007 Page 3 of 610 REJ09B0131-0600
Section 1 Overview
Peripheral clock: 50 MHz (max.) * Supports power-down modes: Sleep mode Software standby mode * Selection of four types of clock modes (PLL2 x2/x4 and clock/crystal resonator are selectable) Ethernet controller (EtherC): * MAC (Media Access Control) function Data frame assembly/disassembly (frame format conforming to IEEE802.3u) CSMA/CD link management (collision prevention and collision processing) CRC processing On-chip FIFOs (256 bytes (SH7618) and 512 bytes (SH7618A), each for transmit/receive operation) Full-duplex transmit/receive support Short frame/long frame detectable * Conforms to the MII (Media Independent Interface) standard Conversion from 8-bit stream data in MAC layer to MII nibble (4-bit) stream Station management (STA function) 18 TTL-level signals 10/100 Mbps transfer rate adjustable * Magic PacketTM* (WOL (Wake-On-LAN) output) Ethernet controller DMAC (EDMAC): * * * * * CPU load reduced with the descriptor management method For transferring from EtherC receive FIFO to receive buffer x 1 channel For transferring from transmit buffer to EtherC transmit FIFO x 1 channel 16-byte burst transfer improves the efficiency of system bus Supports single frame and multiple buffer
Host interface (HIF): * 1 kbyte x 2 banks: in total 2-kbyte buffer RAM * The buffer RAM and the external device are connected in parallel via 16 data pins * The buffer RAM and the CPU of this LSI are connected in parallel via internal bus
Rev. 6.00 Jun. 12, 2007 Page 4 of 610 REJ09B0131-0600
Section 1 Overview
* The external device can access the desired register after the register index has been specified. (However, when the buffer RAM is accessed successively, the address is updated automatically.) * Selection of endian mode * Interrupt requested to the external device * Internal interrupt requested to the CPU of this LSI * Booting from the buffer RAM is enabled if the external device has stored the instruction code in the buffer RAM Compare match timer (CMT): * 16-bit counter * Generates compare match interrupts * Two channels Serial communication interface with FIFO (SCIF): * * * * * * Synchronous and asynchronous modes 16 bytes each for transmit/receive FIFO High-speed UART The UART supports FIFO stop and FIFO trigger Flow control enabled (channel 0 and channel 1 only) Three channels
I/O ports: * 78 general input/output pins * Input or output can be set per bit within the input/output common port Package: * BP1313-176 (0.8 pitch) Power supply voltage: * I/O: 3.0 to 3.6 V Internal: 1.50.1 V (Two power sources are externally provided.) Note: * Magic PacketTM is the registered trademark of Advance Micro Devices, Inc.
Rev. 6.00 Jun. 12, 2007 Page 5 of 610 REJ09B0131-0600
Section 1 Overview
1.2
Block Diagram
Figure 1.1 is a block diagram of this LSI.
SuperH CPU core User break controller (UBC) CPU bus (I clock)
Cache access controller (CCN)
Cache memory 4 kbytes (SH7618) 16 kbytes (SH7618A)
U memory 4 kbytes
Internal bus (B clock)
Bus state controller (BSC)
Peripheral bus controller
Transmit FIFO (256 bytes (SH7618)) Ethernet controller direct memory access controller (E-DMAC) Receive FIFO (512 bytes (SH7618A))
Ethernet controller (EtherC)
External bus
Peripheral bus (P clock)
I/O port, Pin function controller (PFC)
1-kbyte SRAM
Host interface (HIF)
Serial communication interface with FIFO (SCIF)*1
Compare match timer (CMT)*2
User debugging interface (H-UDI)
Interrupt controller (INTC)
Powerdown mode control
Watchdog timer (WDT)
Clock pulse generator (CPG)
Notes: 1. SCIF includes three channels. 2. CMT includes two channels.
Figure 1.1 Block Diagram
Rev. 6.00 Jun. 12, 2007 Page 6 of 610 REJ09B0131-0600
Section 1 Overview
1.3
1
Pin Assignments
2 3 4 5 6 7 8 9
RD
10
11
12
13
14
15
A
VccQ
PA25
PA22
Vcc
PA18
PB08
VccQ
PB05
A13
Vss
A09
A06
VssQ
VccQ
A
B
VssQ
PD7
PA24
Vss
PA19
PA16
VssQ
PB06
PB11
A14
Vcc
A07
A04
A02
A01
B
C
PD3
PD5
PD6
PA21
PA17
PB07
PB09
PB00
PB13
A12
A10
A05
A03
A00
PB04
C
D
PD0
PD2
PD4
PA23
PA20
PB10
PB01
CS0
A15
A11
A08
PB12
PB03
Vss
Vcc
D
E
Vss
Vcc
PE08
PD1
PB02
WE0/ DQMLL
RD/(WR)
WE1/ DQMLU/ WE
E
F
PE22
PE21
PE23
PE24
D09
D08
VssQ
VccQ
F
G
PE18
PE17
PE19
PE20
Vcc
Vss
D10
D11
G
H
PE16
PE15
Vss
Vcc
BP1313-176 (Top view)
D15
D14
D12
D13
H
J
PE12
PE11
PE13
PE14
Vcc
Vss
MD2
CKIO
J
K
VccQ
VssQ
PE09
PE10
D04
D05
D07
D06
K
L
PE06
PE05
PE07
PE04
D00
D01
D02
D03
L
M
Vcc
Vss
PE03
PE01
PC02
PC18
PC05
Vcc
PC19
MD5
TRST
VccQ
Vcc(PLL2)
VssQ
VccQ
M
N
PE02
PE00
PC09
PC08
PC10
PC11
PC06
Vss
TESTOUT
TCK
TDO
ASEMD Vcc(PLL1) Vss(PLL2)
MD1
N
P
PC17
PC16
PC15
PC01
Vcc
PC04
PC12
PC20
MD3
VssQ
TMS
NMI
EXTAL
VssQ
Vss(PLL1)
P
R
VccQ
VssQ
PC00
PC03
Vss
PC13
PC07
PC14
CK_PHY 9
VccQ
TDI
RES
TESTMD
XTAL
MD0
R
1
2
3
4
5
6
7
8
10
11
12
13
14
15
Figure 1.2 Pin Assignments
Rev. 6.00 Jun. 12, 2007 Page 7 of 610 REJ09B0131-0600
Section 1 Overview
1.4
Table 1.1
Classification Power supply
Pin Functions
Pin Functions
Abbr. Vcc I/O Input Pin Name Description
Power Supply Power supply for the internal logic of this LSI. All the Vcc pins must be connected to the system power supply. This LSI does not operate correctly if there is a pin left open. Ground Ground pins. All the Vss pins must be connected to the system power supply (0 V). This LSI does not operate correctly if there is a pin left open.
Vss
Input
VccQ
Input
Power Supply Power supply for input/output pins. All the VccQ pins must be connected to the system power supply. This LSI does not operate correctly if there is a pin left open. Ground Ground pins. All the VssQ pins must be connected to the system power supply (0 V). This LSI does not operate correctly if there is a pin left open.
VssQ
Input
Clock
Vcc (PLL1) Vss (PLL1) Vcc (PLL2) Vss (PLL2) EXTAL
Input Input Input Input Input
Power Supply Power supply pin for the on-chip PLL1 oscillator for PLL1 Ground for PLL1 Ground pin for the on-chip PLL1 oscillator
Power Supply Power supply pin for the on-chip PLL2 oscillator for PLL2 Ground for PLL2 Ground pin for the on-chip PLL2 oscillator
External Clock Connects to a crystal resonator. An external clock is also input on this pin. For details on connection of an external clock, see section 8, Clock Pulse Generator (CPG). Connects to a crystal resonator.
XTAL CKIO CK_PHY
Output Crystal
Output System Clock Supplies the system clock to external devices. Output PHY Clock Mode Setting Supplies the clock for external IEEE802.3-PHY. Sets operating mode. The signal levels of these pins must not be changed during operation. Pins MD2 to MD0 are used for setting clock mode, pin MD3 is for setting bus width mode for area 0, and pin MD5 is for setting endian.
Input Operating MD5, MD3 to MD0 mode control
Rev. 6.00 Jun. 12, 2007 Page 8 of 610 REJ09B0131-0600
Section 1 Overview
Classification System control Interrupt
Abbr. RES NMI IRQ7 to IRQ0
I/O Input Input Input
Pin Name Power-On Reset
Description This LSI enters the reset state when this signal goes low.
Non-Maskable Non-maskable interrupt request signal. When this pin is not Interrupt in use, this signal must be fixed high. Interrupt Maskable interrupt request pins. Request 7 to 0 Level-input or edge-input detection can be selected. When the edge-input detection is selected, the rising or falling edge can also be selected. Outputs addresses. 16-bit bidirectional bus
Address bus Data bus Bus control
A25 to A0 D15 to D0
Output Address Bus Input/ Data Bus output
CS0, CS3, Output Chip Select 0, Chip select signals for external memory and devices. CS4, CS5B, 3, 4, 5B, 6B CS6B RD RD/WR BS WE1 WE0 WAIT RAS CAS CKE DQMLU DQMLL CE1A Output Read Output Read/Write Output Bus Cycle Start Output Upper Side Write Output Lower Side Write Input Wait Indicates that data is read from an external device. Read/write signal Indicates start of a bus cycle. Indicates that bits 15 to 8 of data of external memory or devices are written to. Indicates that bits 7 to 0 of data of external memory or devices are written to. Input pin used to insert wait cycles when accessing the external space Connects to the RAS pin of SDRAM. Connects to the CAS pin of SDRAM. Connects to the CKE pin of SDRAM. Selects bits 15 to 8 of SDRAM data bus. Selects bits 7 to 0 of SDRAM data bus.
Output RAS Output CAS Output Clock Enable Output Upper Side Select Output Lower Side Select
Output PCMCIA Card Chip enable for PCMCIA allocated to area 5 Select Lower Side
Rev. 6.00 Jun. 12, 2007 Page 9 of 610 REJ09B0131-0600
Section 1 Overview
Classification Bus control
Abbr. CE1B
I/O
Pin Name
Description
Output PCMCIA Card Chip enable for PCMCIA allocated to area 6 Select Lower Side Output PCMCIA Card Chip enable for PCMCIA allocated to area 5 Select Upper Side Output PCMCIA Card Chip enable for PCMCIA allocated to area 6 Select Upper Side Output PCMCIA I/O Write Strobe Output PCMCIA I/O Read Strobe Connects to the PCMCIA I/O write strobe pin. Connects to the PCMCIA I/O read strobe pin.
CE2A
CE2B
ICIOWR ICIORD WE
Connects to the PCMCIA memory write strobe. Output PCMCIA Memory Write Strobe Input PCMCIA Dynamic Bus Sizing In little endian mode, this signal indicates 16-bit bus width of PCMCIA. In big endian mode, fix this pin low.
IOIS16
Ethernet controller
CRS COL
Input Input
Carrier Sense Carrier sense pin Collision Collision detect pin
MII_TXD3 to Output Transmit Data 4-bit transmit data pins MII_TXD0 TX_EN TX_CLK TX_ER Output Transmit Enable Input Transmit Clock Indicates that transmit data is on pins MII_TXD3 to MII_TXD0. Timing reference input for the TX_EN, TX_ER, and MII_TXD3 to MII_TXD0 pins
Output Transmit Error Informs PHY LSI of an error during transmission. Receive Data Receive Data Valid 4-bit receive data pins Indicates that valid receive data is on pins MII_RXD3 to MII_RXD0.
MII_RXD3 to Input MII_RXD0 RX_DV RX_CLK RX_ER Input Input Input
Receive Clock Timing reference input for the RX_DV, RX_ER, and MII_RXD3 to MII_RXD0 pins Receive Error Pin for detection of an error during reception
Rev. 6.00 Jun. 12, 2007 Page 10 of 610 REJ09B0131-0600
Section 1 Overview
Classification Ethernet controller
Abbr. MDC MDIO WOL LNKSTA EXOUT
I/O
Pin Name
Description Timing reference input for transfer information on the MDIO pin Bidirectional pin for management information transfer
TM
Output Management Clock Input/ Management output Data I/O
Output MAGIC Packet Indicates that a Magic Packet * has received. Receive Input Link Status Input pin for a link state from a PHY LSI.
Output General output Output pin to external devices Output Transmit Data Transmit data pins Input Receive Data Receive data pins Clock input pins Modem control pin. Supported only by SCIF0 and SCIF1. Modem control pin. Supported only by SCIF0 and SCIF1. Address, data, and command input/output pins for the HIF. Chip select input for the HIF Controls the access type switching for the HIF. Write strobe signal Read strobe signal Interrupt request to external devices by the HIF Specifies HIF boot mode. Requests DMAC transfer for the HIFRAM to external devices. Indicates that a reset of the HIF has been cleared in this LSI and the HIF is ready for accesses to it. HIF pins other than this pin are enabled by driving this pin high.
Serial communic ations interface with FIFO
TXD2 to TXD0 RXD2 to RXD0 SCK2 to SCK0 RTS1 and RTS0 CTS1 and CTS0
Input/ Serial clock output Output Transmit Request Input Transmit Enable
Host interface
HIFD15 to HIFD0 HIFCS HIFRS HIFWR HIFRD HIFINT HIFMD HIFDREQ
Input/ HIF Data Bus output Input Input Input Input HIF Chip Select HIF Register Select HIF Write HIF Read
Output HIF Interrupt Input HIF Mode
Output HIF DMAC Transfer Request Output HIF Boot Ready Input HIF Pin Enable
HIFRDY HIFEBL
Rev. 6.00 Jun. 12, 2007 Page 11 of 610 REJ09B0131-0600
Section 1 Overview
Classification
Abbr.
I/O Input Input Input
Pin Name Test Clock Test Mode Select Test Data Input
Description Test clock input pin Input pin for test mode select signal Serial input pin for an instruction and data Serial output pin for an instruction and data Input pin for initialization Pins for 10-bit general input/output port Pins for 14-bit general input/output port Pins for 21-bit general input/output port Pins for 8-bit general input/output port Pins for 25-bit general input/output port Specifies ASE mode. This LSI enters ASE mode when this signal goes low and normal mode when this pin goes high. In ASE mode, functions for the emulator are available.
TCK User debugging TMS interface (H-UDI) TDI TDO TRST I/O port PA25 to PA16 PB13 to PB00 PC20 to PC00 PD07 to PD00 PE24 to PE00 Emulator interface ASEMD
Output Test Data Output Input Test Reset
Input/ General port output Input/ General port output Input/ General port output Input/ General port output Input/ General port output Input ASE Mode
Test Mode TESTMD
Input
Test Mode
Specifies test mode. This LSI enters test mode when this signal goes low. Fix this signal high.
TESTOUT
Output Test Output
TM
Output pin for testing. This pin should be open.
Note:
*
Magic Packet
is the trademark of Advanced Micro Devices, Inc.
Rev. 6.00 Jun. 12, 2007 Page 12 of 610 REJ09B0131-0600
Section 1 Overview
Table 1.2
Pin No. A1 A2 A3 A4 A5 A6 A7 A8 A9 A10 A11 A12 A13 A14 A15 B1 B2 B3 B4 B5 B6 B7 B8 B9 B10 B11 B12 B13 B14 B15
Pin Features
Pin Name VCCQ PA25/A25 PA22/A22 VCC PA18/A18 PB08/(CS6B/CE1B) VCCQ PB05/ICIORD RD A13 VSS A09 A06 VSSQ VCCQ VSSQ PD7/IRQ7/SCK2 PA24/A24 VSS PA19/A19 PA16/A16 VSSQ PB06/ICIOWR PB11/CS4 A14 VCC A07 A04 A02 A01 I/O Features Power IO/O IO/O Power IO/O IO/O/O Power IO/O O O Power O O Power Power Power IO/I/IO IO/O Power IO/O IO/O Power IO/O IO/O O Power O O O O
Rev. 6.00 Jun. 12, 2007 Page 13 of 610 REJ09B0131-0600
Section 1 Overview
Pin No. C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 C13 C14 C15 D1 D2 D3 D4 D5 D6 D7 D8 D9 D10 D11 D12 D13 D14 D15 E1 E2
Pin Name PD3/IRQ3/RxD1 PD5/IRQ5/TxD2 PD6/IRQ6/RxD2 PA21/A21 PA17/A17 PB07/CE2B PB09/CE2A PB00/WAIT PB13/BS A12 A10 A05 A03 A00 PB04/RAS PD0/IRQ0 PD2/IRQ2/TxD1 PD4/IRQ4/SCK1 PA23/A23 PA20/A20 PB10/(CS5B/CE1A) PB01/IOIS16 CS0 A15 A11 A08 PB12/CS3 PB03/CAS VSS VCC VSS VCC
I/O Features IO/I/I IO/I/O IO/I/I IO/O IO/O IO/O IO/O IO/I IO/O O O O O O IO/O IO/I IO/I/O IO/I/IO IO/O IO/O IO/O/O IO/I O O O O IO/O IO/O Power Power Power Power
Rev. 6.00 Jun. 12, 2007 Page 14 of 610 REJ09B0131-0600
Section 1 Overview
Pin No. E3 E4 E12 E13 E14 E15 F1 F2 F3 F4 F12 F13 F14 F15 G1 G2 G3 G4 G12 G13 G14 G15 H1 H2 H3 H4 H12 H13 H14 H15 J1 J2
Pin Name PE08/HIFCS PD1/IRQ1 PB02/CKE (WE0/DQMLL) RD/(WR) (WE1/DQMLU/WE) PE22/HIFD13/CTS0 PE21/HIFD12/RTS0 PE23/HIFD14/RTS1 PE24/HIFD15/CTS1 D09 D08 VSSQ VCCQ PE18/HIFD09/TxD1 PE17/HIFD08/SCK0 PE19/HIFD10/RxD1 PE20/HIFD11/SCK1 VCC VSS D10 D11 PE16/HIFD07/RxD0 PE15/HIFD06/TxD0 VSS VCC D15 D14 D12 D13 PE12/HIFD03 PE11/HIFD02
I/O Features IO/I IO/I IO/O O/O O O/O/O IO/IO/I IO/IO/O IO/IO/O IO/IO/I IO IO Power Power IO/IO/O IO/IO/IO IO/IO/I IO/IO/IO Power Power IO IO IO/IO/I IO/IO/O Power Power IO IO IO IO IO/IO IO/IO
Rev. 6.00 Jun. 12, 2007 Page 15 of 610 REJ09B0131-0600
Section 1 Overview
Pin No. J3 J4 J12 J13 J14 J15 K1 K2 K3 K4 K12 K13 K14 K15 L1 L2 L3 L4 L12 L13 L14 L15 M1 M2 M3 M4 M5 M6 M7 M8 M9 M10
Pin Name PE13/HIFD04 PE14/HIFD05 VCC VSS MD2 CKIO VCCQ VSSQ PE09/HIFD00 PE10/HIFD01 D04 D05 D07 D06 PE06/HIFWR PE05/HIFRD PE07/HIFRS PE04/HIFINT D00 D01 D02 D03 VCC VSS PE03/HIFMD PE01/HIFRDY PC02/MII_RXD2 PC18/LNKSTA PC05/MII_TXD1 VCC PC19/EXOUT MD5
I/O Features IO/IO IO/IO Power Power I IO Power Power IO/IO IO/IO IO IO IO IO IO/I IO/I IO/I IO/O IO IO IO IO Power Power IO/I IO/O IO/I IO/I IO/O Power IO/O I
Rev. 6.00 Jun. 12, 2007 Page 16 of 610 REJ09B0131-0600
Section 1 Overview
Pin No. M11 M12 M13 M14 M15 N1 N2 N3 N4 N5 N6 N7 N8 N9 N10 N11 N12 N13 N14 N15 P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12
Pin Name TRST VCCQ VCC (PLL2) VSSQ VCCQ PE02/HIFDREQ PE00/HIFEBL PC09/RX_ER PC08/RX_DV PC10/RX_CLK PC11/TX_ER PC06/MII_TXD2 VSS TESTOUT TCK TDO ASEMD VCC (PLL1) VSS (PLL2) MD1 PC17/MDC PC16/MDIO PC15/CRS PC01/MII_RXD1 VCC PC04/MII_TXD0 PC12/TX_EN PC20/WOL MD3 VSSQ TMS NMI
I/O Features I Power Power Power Power IO/O IO/I IO/I IO/I IO/I IO/O IO/O Power O I O I Power Power I IO/O IO/IO IO/I IO/I Power IO/O IO/O IO/O I Power I I
Rev. 6.00 Jun. 12, 2007 Page 17 of 610 REJ09B0131-0600
Section 1 Overview
Pin No. P13 P14 P15 R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15
Pin Name EXTAL VSSQ VSS (PLL1) VCCQ VSSQ PC00/MII_RXD0 PC03/MII_RXD3 VSS PC13/TX_CLK PC07/MII_TXD3 PC14/COL CK_PHY VCCQ TDI RES TESTMD XTAL MD0
I/O Features I Power Power Power Power IO/I IO/I Power IO/I IO/O IO/I O Power I I I O I
Rev. 6.00 Jun. 12, 2007 Page 18 of 610 REJ09B0131-0600
Section 2 CPU
Section 2 CPU
2.1 Features
* General registers: 32-bit register x 16 * Basic instructions: 62 * Addressing modes: 11 Register direct (Rn) Register indirect (@Rn) Post-increment register indirect (@Rn+) Pre-decrement register indirect (@-Rn) Register indirect with displacement (@disp:4, Rn) Index register indirect (@R0, Rn) GBR indirect with displacement (@disp:8, GBR) Index GBR indirect (@R0, GBR) PC relative with displacement (@disp:8, PC) PC relative (disp:8/disp:12/Rn) Immediate (#imm:8)
2.2
Register Configuration
There are three types of registers: general registers (32-bit x 16), control registers (32-bit x 3), and system registers (32-bit x 4).
CPUS200C_000020020700
Rev. 6.00 Jun. 12, 2007 Page 19 of 610 REJ09B0131-0600
Section 2 CPU
General register (Rn)
31 R0*1 R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15, SP (hardware stack pointer)0*2 0
Status register (SR)
31 9 8765 43 210
M Q I3 I2 I1 I0 ST
Global base register (GBR)
31
GBR
0
Vector base register (VBR)
31
VBR
0
Multiply and accumulate register (MAC)
31
MACH MACL
0
Procedure register (PR)
31
PR
0
Program counter (PC)
31
PC
0
Notes: 1. R0 can be used as an index register in index register indirect or index GBR indirect addressing mode. For some instructions, only R0 is used as the source or destination register. 2. R15 is used as a hardware stack pointer during exception handling.
Figure 2.1 CPU Internal Register Configuration
Rev. 6.00 Jun. 12, 2007 Page 20 of 610 REJ09B0131-0600
Section 2 CPU
2.2.1
General Registers (Rn)
There are sixteen 32-bit general registers (Rn), designated R0 to R15. The general registers are used for data processing and address calculation. R0 is also used as an index register. With a number of instructions, R0 is the only register that can be used. R15 is used as a hardware stack pointer (SP). In exception handling, R15 is used for accessing the stack to save or restore the status register (SR) and program counter (PC) values. 2.2.2 Control Registers
There are three 32-bit control registers, designated status register (SR), global base register (GBR), and vector base register (VBR). SR indicates a processing state. GBR is used as a base address in GBR indirect addressing mode for data transfer of on-chip peripheral module registers. VBR is used as a base address of the exception handling (including interrupts) vector table. * Status register (SR)
Bit 31 to 10 Bit name Default All 0 Read/ Write R/W Description Reserved These bits are always read as 0. The write value should always be 0. 9 8 7 6 5 4 3, 2 M Q I3 I2 I1 I0 Undefined Undefined 1 1 1 1 All 0 R/W R/W R/W R/W R/W R/W R/W Reserved These bits are always read as 0. The write value should always be 0. 1 S Undefined R/W S Used by the multiply and accumulate instruction. Used by the DIV0U, DIV0S, and DIV1 instructions. Used by the DIV0U, DIV0S, and DIV1 instructions. Interrupt Mask
Rev. 6.00 Jun. 12, 2007 Page 21 of 610 REJ09B0131-0600
Section 2 CPU
Bit 0
Bit name T
Default Undefined
Read/ Write R/W
Description T Indicates true (1) or false (0) in the following instructions: MOVT, CMP/cond, TAS, TST, BT (BT/S), BF (BF/S), SETT, CLRT Indicates carry, borrow, overflow, or underflow in the following instructions: ADDV, ADDC, SUBV, SUBC, NEGC, DIV0U, DIV0S, DIV1, SHAR, SHAL, SHLR, SHLL, ROTR, ROTL, ROTCR, ROTCL
* Global-base register (GBR) This register indicates a base address in GBR indirect addressing mode. The GBR indirect addressing mode is used for data transfer of the on-chip peripheral module registers and logic operations. * Vector-base register (VBR) This register indicates the base address of the exception handling vector table. 2.2.3 System Registers
There are four 32-bit system registers, designated two multiply and accumulate registers (MACH and MACL), a procedure register (PR), and program counter (PC). * Multiply and accumulate registers (MAC) This register stores the results of multiplication and multiply-and-accumulate operation. * Procedure register (PR) This register stores the return-destination address from subroutine procedures. * Program counter (PC) The PC indicates the point which is four bytes (two instructions) after the current execution instruction.
Rev. 6.00 Jun. 12, 2007 Page 22 of 610 REJ09B0131-0600
Section 2 CPU
2.2.4
Initial Values of Registers
Table 2.1 lists the initial values of registers after a reset. Table 2.1 Initial Values of Registers
Register R0 to R14 R15 (SP) Control register SR Default Undefined SP value set in the exception handling vector table I3 to I0: 1111 (H'F) Reserved bits: 0 Other bits: Undefined GBR VBR System register MACH, MACL, PR PC Undefined H'00000000 Undefined PC value set in the exception handling vector table
Type of register General register
Rev. 6.00 Jun. 12, 2007 Page 23 of 610 REJ09B0131-0600
Section 2 CPU
2.3
2.3.1
Data Formats
Register Data Format
The size of register operands is always longwords (32 bits). When loading byte (8 bits) or word (16 bits) data in memory into a register, the data is sign-extended to longword and stored in the register.
31 Longword
0
Figure 2.2 Register Data Format 2.3.2 Memory Data Formats
Memory data formats are classified into byte, word, and longword. Byte data can be accessed from any address. If word data starting from boundary other than 2n or longword data starting from a boundary other than 4n is accessed, an address error will occur. In such cases, the data accessed cannot be guaranteed. See figure 2.3.
Address A + 1 Address A 31 Address A Address A + 4 Address A + 8 Byte 0 23 Byte 1 Address A + 3 31 Byte 3 Address A + 10 Address A + 11 0 Byte 3 23 Byte 2 Address A + 8
Address A + 2 15 Byte 2 7
Address A + 9 15 Byte 1 7 Byte 0 0 Address A + 8 Address A + 4 Address A
Word 0 Longword Big endian
Word 1
Word 1 Longword Little endian
Word 0
Figure 2.3 Memory Data Format Either big endian and little endian formats can be selected according to the mode pin setting at a reset. For details on mode pin settings, see section 7, Bus State Controller (BSC).
Rev. 6.00 Jun. 12, 2007 Page 24 of 610 REJ09B0131-0600
Section 2 CPU
2.3.3
Immediate Data Formats
Immediate data of eight bits is placed in the instruction code. For the MOV, ADD, and CMP/EQ instructions, the immediate data is sign-extended to longword and then calculated. For the TST, AND, OR, and XOR instructions, the immediate data is zeroextended to longword and then calculated. Thus, if the immediate data is used for the AND instruction, the upper 24 bits in the destination register are always cleared. The immediate data of word or longword is not placed in the instruction code. It is placed in a table in memory. The table in memory is accessed by the MOV immediate data instruction in PC relative addressing mode with displacement.
2.4
2.4.1
Features of Instructions
RISC Type
The instructions are RISC-type instructions with the following features: Fixed 16-Bit Length: All instructions have a fixed length of 16 bits. This improves program code efficiency. One Instruction per Cycle: Since pipelining is used, basic instructions can be executed in one cycle. One cycle is 25ns with 40 MHz operation. Data Size: The basic data size for operations is longword. Byte, word, or longword can be selected as the memory access size. Byte or word data in memory is sign-extended to longword and then calculated. Immediate data is sign-extended to longword for arithmetic operations or zero-extended to longword size for logical operations. Table 2.2 Word Data Sign Extension
Description Sign-extended to 32 bits, R1 becomes H'00001234, and is then operated on by the ADD instruction. Example of Other CPUs ADD.W #H'1234,R0
CPU in this LSI MOV.W ADD @(disp,PC),R1 R1,R0 ........ .DATA.W H'1234
Note: Immediate data is accessed by @(disp,PC).
Rev. 6.00 Jun. 12, 2007 Page 25 of 610 REJ09B0131-0600
Section 2 CPU
Load/Store Architecture: Basic operations are executed between registers. In operations involving memory, data is first loaded into a register (load/store architecture). However, bit manipulation instructions such as AND are executed directly in memory. Delayed Branching: Unconditional branch instructions means the delayed branch instructions. With a delayed branch instruction, the branch is made after execution of the instruction immediately following the delayed branch instruction. This minimizes disruption of the pipeline when a branch is made. The conditional branch instructions have two types of instructions: conditional branch instructions and delayed branch instructions. Table 2.3 Delayed Branch Instructions
Description ADD is executed before branch to TRGET. Example of Other CPUs ADD.W R1,R0 BRA TRGET
CPU in this LSI BRA ADD TRGET R1,R0
Multiply/Multiply-and-Accumulate Operations: A 16 x 16 32 multiply operation is executed in one to two cycles, and a 16 x 16 + 64 64 multiply-and-accumulate operation in two to three cycles. A 32 x 32 64 multiply operation and a 32 x 32 + 64 64 multiply-andaccumulate operation are each executed in two to four cycles. T Bit: The result of a comparison is indicated by the T bit in SR, and a conditional branch is performed according to whether the result is True or False. Processing speed has been improved by keeping the number of instructions that modify the T bit to a minimum. Table 2.4 T Bit
Description When R0 R1, the T bit is set. Example of Other CPUs CMP.W R1,R0 TRGET0 TRGET1 TRGET
CPU in this LSI CMP/GE BT BF ADD CMP/EQ BT R1,R0 TRGET0 TRGET1 #-1,R0 #0,R0 TRGET
When R0 R1, a branch is made to TRGET0. BGE When R0 < R1, a branch is made to TRGET1. BLT The T bit is not changed by ADD. When R0 = 0, the T bit is set. A branch is made when R0 = 0. BEQ
SUB.W #1,R0
Rev. 6.00 Jun. 12, 2007 Page 26 of 610 REJ09B0131-0600
Section 2 CPU
Immediate Data: 8-bit immediate data is placed in the instruction code. Word and longword immediate data is not placed in the instruction code. It is placed in a table in memory. The table in memory is accessed with the MOV immediate data instruction using PC relative addressing mode with displacement. Table 2.5
Type 8-bit immediate 16-bit immediate
Access to Immediate Data
This LSI's CPU MOV #H'12,R0 ........ .DATA.W H'1234 Example of Other CPU MOV.B #H'12,R0
MOV.W @(disp,PC),R0
MOV.W #H'1234,R0
32-bit immediate
MOV.L
@(disp,PC),R0 ........
MOV.L #H'12345678, R0
.DATA.L H'12345678 Note: Immediate data is accessed by @(disp,PC).
Absolute Addresses: When data is accessed by absolute address, place the absolute address value in a table in memory beforehand. The absolute address value is transferred to a register using the method whereby immediate data is loaded when an instruction is executed, and the data is accessed using the register indirect addressing mode. Table 2.6
Type Absolute address
Access to Absolute Address
CPU in this LSI MOV.L MOV.B @(disp,PC),R1 @R1,R0 ........ .DATA.L H'12345678 Example of Other CPUs MOV.B @H'12345678, R0
Note: Immediate data is referenced by @(disp,PC).
16-Bit/32-Bit Displacement: When data is accessed using the 16- or 32-bit displacement addressing mode, the displacement value is placed in a table in memory beforehand. Using the method whereby immediate data is loaded when an instruction is executed, this value is transferred to a register and the data is accessed using index register indirect addressing mode.
Rev. 6.00 Jun. 12, 2007 Page 27 of 610 REJ09B0131-0600
Section 2 CPU
Table 2.7
Type
Access with Displacement
CPU in this LSI MOV.W @(disp,PC),R0 MOV.W @(R0,R1),R2 ........ .DATA.W H'1234 Example of Other CPUs MOV.W @(H'1234,R1), R2
16-bit displacement
Note: Immediate data is referenced by @(disp,PC).
2.4.2
Addressing Modes
Table 2.8 lists addressing modes and effective address calculation methods. Table 2.8
Addressing Mode Register direct Register indirect Register indirect with post-increment
Addressing Modes and Effective Addresses
Instruction Format Effective Address Calculation Method Rn @Rn @Rn+ Effective address is register Rn. (Operand is register Rn contents.) Effective address is register Rn contents.
Rn Rn
Calculation Formula Rn Rn After instruction execution Byte: Rn + 1 Rn Word: Rn + 2 Rn Longword: Rn + 4 Rn
Effective address is register Rn contents. A constant is added to Rn after instruction execution: 1 for a byte operand, 2 for a word operand, 4 for a longword operand.
Rn Rn + 1/2/4 + 1/2/4 Rn
Rev. 6.00 Jun. 12, 2007 Page 28 of 610 REJ09B0131-0600
Section 2 CPU
Addressing Mode Register indirect with pre-decrement
Instruction Format Effective Address Calculation Method @-Rn Effective address is register Rn contents, decremented by a constant beforehand: 1 for a byte operand, 2 for a word operand, 4 for a longword operand.
Rn Rn - 1/2/4
Calculation Formula Byte: Rn - 1 Rn Word: Rn - 2 Rn Longword: Rn - 4 Rn (Instruction executed with Rn after calculation) Byte: Rn + disp Word: Rn + disp x 2 Longword: Rn + disp x 4
1/2/4
Rn - 1/2/4
Register indirect with displacement
@(disp:4, Rn)
Effective address is register Rn contents with 4-bit displacement disp added. After disp is zero-extended, it is multiplied by 1 (byte), 2 (word), or 4 (longword), according to the operand size.
Rn disp (zero-extended) x 1/2/4
+
Rn + disp x 1/2/4
Index @(R0, Rn) Effective address is sum of register Rn and R0 register indirect contents.
Rn
Rn + R0
+
R0
Rn + R0
GBR indirect with displacement
@(disp:8, GBR)
Effective address is register GBR contents with 8-bit displacement disp added. After disp is zero-extended, it is multiplied by 1 (byte), 2 (word), or 4 (longword), according to the operand size.
GBR disp (zero-extended) x 1/2/4
Byte: GBR + disp Word: GBR + disp x2 Longword: GBR + disp x 4
+
GBR + disp x 1/2/4
Rev. 6.00 Jun. 12, 2007 Page 29 of 610 REJ09B0131-0600
Section 2 CPU
Addressing Mode Index GBR indirect
Instruction Format Effective Address Calculation Method @(R0, GBR) Effective address is sum of register GBR and R0 contents.
GBR
Calculation Formula GBR + R0
+
R0
GBR + R0
PC relative with @(disp:8, displacement PC)
Effective address is PC with 8-bit displacement disp added. After disp is zero-extended, it is multiplied by 2 (word) or 4 (longword), according to the operand size. With a longword operand, the lower 2 bits of PC are masked.
PC & H'FFFFFFFC disp (zero-extended) * PC + disp x 2 or PC& H'FFFFFFFC + disp x 4 *With longword operand
Word: PC + disp x2 Longword: PC&H'FFFFFFFC + disp x 4
+
x
2/4
PC relative
disp:8
Effective address is PC with 8-bit displacement disp added after being sign-extended and multiplied by 2.
PC disp (sign-extended) x 2
PC + disp x 2
+
PC + disp x 2
disp:12
Effective address is PC with 12-bit displacement PC + disp x 2 disp added after being sign-extended and multiplied by 2.
PC disp (sign-extended) x 2
+
PC + disp x 2
Rev. 6.00 Jun. 12, 2007 Page 30 of 610 REJ09B0131-0600
Section 2 CPU
Addressing Mode PC relative
Instruction Format Effective Address Calculation Method Rn Effective address is sum of PC and Rn.
PC
Calculation Formula PC + Rn
+
Rn
PC + Rn
Immediate
#imm:8 #imm:8 #imm:8
8-bit immediate data imm of TST, AND, OR, or XOR instruction is zero-extended. 8-bit immediate data imm of MOV, ADD, or CMP/EQ instruction is sign-extended. 8-bit immediate data imm of TRAPA instruction is zero-extended and multiplied by 4.

2.4.3
Instruction Formats
This section describes the instruction formats, and the meaning of the source and destination operands. The meaning of the operands depends on the instruction code. The following symbols are used in the table. xxxx: Instruction code mmmm: Source register nnnn: Destination register iiii: Immediate data dddd: Displacement
Rev. 6.00 Jun. 12, 2007 Page 31 of 610 REJ09B0131-0600
Section 2 CPU
Table 2.9
Instruction Formats
Source Operand Destination Operand Sample Instruction NOP
Instruction Format 0 type
15 0 xxxx xxxx xxxx xxxx
n type
15 0 xxxx nnnn xxxx xxxx
nnnn: register direct
MOVT Rn STS MACH,Rn
Control register or nnnn: register system register direct
Control register or nnnn: preSTC.L SR,@-Rn system register decrement register indirect m type
15 0 xxxx mmmm xxxx xxxx
mmmm: register direct
Control register or LDC Rm,SR system register
mmmm: postControl register or LDC.L @Rm+,SR increment register system register indirect mmmm: register indirect PC relative using Rm JMP @Rm BRAF Rm
Rev. 6.00 Jun. 12, 2007 Page 32 of 610 REJ09B0131-0600
Section 2 CPU
Instruction Format nm type
15 0 xxxx nnnn mmmm xxxx
Source Operand mmmm: register direct mmmm: register direct
Destination Operand nnnn: register direct nnnn: register indirect
Sample Instruction ADD Rm,Rn
MOV.L Rm,@Rn MAC.W @Rm+,@Rn+
mmmm: postMACH, MACL increment register indirect (multiplyand-accumulate operation) nnnn: * postincrement register indirect (multiplyand-accumulate operation) mmmm: postnnnn: register increment register direct indirect mmmm: register direct mmmm: register direct md type
15 0 xxxx xxxx mmmm dddd
MOV.L @Rm+,Rn
nnnn: preMOV.L Rm,@-Rn decrement register indirect nnnn: index register indirect MOV.L Rm,@(R0,Rn)
mmmmdddd: register indirect with displacement
R0 (register direct) MOV.B @(disp,Rm),R0
nd4 type
15 0 xxxx xxxx nnnn dddd
R0 (register direct) nnnndddd: register indirect with displacement mmmm: register direct mmmmdddd: register indirect with displacement nnnndddd: register indirect with displacement nnnn: register direct
MOV.B R0,@(disp,Rn)
nmd type
15 0 xxxx nnnn mmmm dddd
MOV.L Rm,@(disp,Rn)
MOV.L @(disp,Rm),Rn
Rev. 6.00 Jun. 12, 2007 Page 33 of 610 REJ09B0131-0600
Section 2 CPU
Instruction Format d type
15 0 xxxx xxxx dddd dddd
Source Operand dddddddd: GBR indirect with displacement
Destination Operand
Sample Instruction
R0 (register direct) MOV.L @(disp,GBR),R0 MOV.L R0,@(disp,GBR)
R0 (register direct) dddddddd: GBR indirect with displacement dddddddd: PC relative with displacement d12 type
15 0 xxxx dddd dddd dddd
R0 (register direct) MOVA @(disp,PC),R0
dddddddd: PC relative dddddddddddd: PC relative nnnn: register direct
BF label BRA label (label=disp+PC) MOV.L @(disp,PC),Rn
nd8 type
15 0 xxxx nnnn dddd dddd
dddddddd: PC relative with displacement iiiiiiii: immediate iiiiiiii: immediate iiiiiiii: immediate
i type
15 0 xxxx xxxx iiii iiii
Index GBR indirect AND.B #imm,@(R0,GBR) R0 (register direct) AND #imm,R0 nnnn: register direct TRAPA #imm ADD #imm,Rn
ni type
15 0 xxxx nnnn iiii iiii
iiiiiiii: immediate
Note:
*
In multiply and accumulate instructions, nnnn is the source register.
Rev. 6.00 Jun. 12, 2007 Page 34 of 610 REJ09B0131-0600
Section 2 CPU
2.5
2.5.1
Instruction Set
Instruction Set by Type
Table 2.10 lists the instructions classified by type. Table 2.10 Instruction Types
Type Data transfer instructions Kinds of Instruction 5 Op Code MOV Function Data transfer Immediate data transfer Peripheral module data transfer Structure data transfer MOVA MOVT SWAP XTRCT Arithmetic operation instructions 21 ADD ADDC ADDV CMP/cond DIV1 DIV0S DIV0U DMULS DMULU DT EXTS EXTU MAC MUL Effective address transfer T bit transfer Upper/lower swap Extraction of middle of linked registers Binary addition Binary addition with carry Binary addition with overflow Comparison Division Signed division initialization Unsigned division initialization Signed double-precision multiplication Unsigned double-precision multiplication Decrement and test Sign extension Zero extension Multiply-and-accumulate, doubleprecision multiply-and-accumulate Double-precision multiplication 33 Number of Instructions 39
Rev. 6.00 Jun. 12, 2007 Page 35 of 610 REJ09B0131-0600
Section 2 CPU
Type Arithmetic operation instructions
Kinds of Instruction 21
Op Code MULS MULU NEG NEGC SUB SUBC SUBV
Function Signed multiplication Unsigned multiplication Sign inversion Sign inversion with borrow Binary subtraction Binary subtraction with carry Binary subtraction with underflow Logical AND Bit inversion Logical OR Memory test and bit setting T bit setting for logical AND Exclusive logical OR 1-bit left shift 1-bit right shift 1-bit left shift with T bit 1-bit right shift with T bit Arithmetic 1-bit left shift Arithmetic 1-bit right shift Logical 1-bit left shift Logical n-bit left shift Logical 1-bit right shift Logical n-bit right shift
Number of Instructions 33
Logic operation instructions
6
AND NOT OR TAS TST XOR
14
Shift instructions
10
ROTL ROTR ROTCL ROTCR SHAL SHAR SHLL SHLLn SHLR SHLRn
14
Rev. 6.00 Jun. 12, 2007 Page 36 of 610 REJ09B0131-0600
Section 2 CPU
Type Branch instructions
Kinds of Instruction 9
Op Code BF BT BRA BRAF BSR BSRF JMP JSR RTS
Function
Number of Instructions
Conditional branch, delayed conditional 11 branch (T = 0) Conditional branch, delayed conditional branch (T = 1) Unconditional branch Unconditional branch Branch to subroutine procedure Branch to subroutine procedure Unconditional branch Branch to subroutine procedure Return from subroutine procedure T bit clear MAC register clear Load into control register Load into system register No operation Return from exception handling T bit setting Transition to power-down mode Store from control register Store from system register Trap exception handling 142 31
System control instructions
11
CLRT CLRMAC LDC LDS NOP RTE SETT SLEEP STC STS TRAPA
Total:
62
Rev. 6.00 Jun. 12, 2007 Page 37 of 610 REJ09B0131-0600
Section 2 CPU
The instruction code, operation, and execution cycles of the instructions are listed in the following tables, classified by type.
Instruction Instruction Code Summary of Operation
Indicates summary of operation.
Execution Cycles T Bit
Value when no Value of T bit after wait cycles are instruction is 1 inserted* executed Explanation of
Indicated by mnemonic. Indicated in MSB LSB order.
Explanation of Symbols Explanation of Symbols Explanation of Symbols mmmm: Source register OP.Sz SRC, DEST OP: Operation code nnnn: Destination Sz: Size register SRC: Source 0000: R0 DEST: Destination 0001: R1 ......... Rm: Source register 1111: R15 Rn: Destination register iiii: Immediate data imm: Immediate data disp: Displacement*
2
Symbols : No change
, : (xx):
Transfer direction Memory operand
M/Q/T: Flag bits in SR &: |: ^: Logical AND of each bit Logical OR of each bit Exclusive logical OR of each bit -: Logical NOT of each bit
dddd: Displacement
<>n: n-bit right shift
Notes: 1. The table shows the minimum number of execution states. In practice, the number of instruction execution states will be increased in cases such as the following: * When there is contention between an instruction fetch and a data access * When the destination register of a load instruction (memory register) is also used by the following instruction 2. Scaled (x1, x2, or x4) according to the instruction operand size, etc. For details, see SH-1/SH-2/SH-DSP Software Manual.
Rev. 6.00 Jun. 12, 2007 Page 38 of 610 REJ09B0131-0600
Section 2 CPU
* Data Transfer Instructions
Instruction MOV #imm,Rn Operation
imm Sign extension Rn extension Rn
Code
1110nnnniiiiiiii
Execution Cycles 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
T Bit
MOV.W @(disp,PC),Rn (disp x 2 + PC) Sign MOV.L @(disp,PC),Rn (disp x 4 + PC) Rn MOV Rm,Rn
Rm Rn Rm (Rn) Rm (Rn) Rm (Rn) (Rm) Sign extension Rn (Rm) Sign extension Rn (Rm) Rn Rn-1 Rn, Rm (Rn) Rn-2 Rn, Rm (Rn) Rn-4 Rn, Rm (Rn) (Rm) Sign extension Rn, Rm + 1 Rm (Rm) Sign extension Rn, Rm + 2 Rm
1001nnnndddddddd
1101nnnndddddddd 0110nnnnmmmm0011 0010nnnnmmmm0000 0010nnnnmmmm0001 0010nnnnmmmm0010 0110nnnnmmmm0000
MOV.B Rm,@Rn MOV.W Rm,@Rn MOV.L Rm,@Rn MOV.B @Rm,Rn MOV.W @Rm,Rn MOV.L @Rm,Rn MOV.B Rm,@-Rn MOV.W Rm,@-Rn MOV.L Rm,@-Rn MOV.B @Rm+,Rn MOV.W @Rm+,Rn MOV.L @Rm+,Rn
0110nnnnmmmm0001
0110nnnnmmmm0010 0010nnnnmmmm0100 0010nnnnmmmm0101 0010nnnnmmmm0110 0110nnnnmmmm0100
0110nnnnmmmm0101
(Rm) Rn,Rm + 4 Rm 0110nnnnmmmm0110 10000000nnnndddd 10000001nnnndddd 0001nnnnmmmmdddd 10000100mmmmdddd
MOV.B R0,@(disp,Rn) R0 (disp + Rn) MOV.W R0,@(disp,Rn) R0 (disp x 2 + Rn) MOV.L Rm,@(disp,Rn) Rm (disp x 4 + Rn) MOV.B @(disp,Rm),R0 (disp + Rm) Sign
extension R0
MOV.W @(disp,Rm),R0 (disp x 2 + Rm) Sign
extension R0
10000101mmmmdddd
MOV.L @(disp,Rm),Rn (disp x 4 + Rm) Rn MOV.B Rm,@(R0,Rn) MOV.W Rm,@(R0,Rn)
Rm (R0 + Rn) Rm (R0 + Rn)
0101nnnnmmmmdddd 0000nnnnmmmm0100 0000nnnnmmmm0101
Rev. 6.00 Jun. 12, 2007 Page 39 of 610 REJ09B0131-0600
Section 2 CPU
Instruction MOV.L Rm,@(R0,Rn) MOV.B @(R0,Rm),Rn MOV.W @(R0,Rm),Rn MOV.L @(R0,Rm),Rn
MOV.B MOV.W MOV.L MOV.B
Operation
Rm (R0 + Rn) (R0 + Rm) Sign extension Rn (R0 + Rm) Sign extension Rn (R0 + Rm) Rn
Code
0000nnnnmmmm0110 0000nnnnmmmm1100
Execution Cycles 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
T Bit
0000nnnnmmmm1101
0000nnnnmmmm1110 11000000dddddddd 11000001dddddddd 11000010dddddddd 11000100dddddddd
R0,@(disp,GBR) R0 (disp + GBR) R0,@(disp,GBR) R0 (disp x 2 + GBR) R0,@(disp,GBR) R0 (disp x 4 + GBR) @(disp,GBR),R0 (disp + GBR) Sign extension R0 @(disp,GBR),R0 (disp x 2 + GBR) Sign extension R0 @(disp,GBR),R0 (disp x 4 + GBR) R0
MOV.W
11000101dddddddd
MOV.L
11000110dddddddd 11000111dddddddd 0000nnnn00101001 0110nnnnmmmm1000
MOVA MOVT
@(disp,PC),R0 disp x 4 + PC R0 Rn
T Rn Rm Swap lowest two bytes Rn Rm Swap two consecutive words Rn Rm: Middle 32 bits of Rn Rn
SWAP.B Rm,Rn SWAP.W Rm,Rn XTRCT Rm,Rn
0110nnnnmmmm1001
0010nnnnmmmm1101
Rev. 6.00 Jun. 12, 2007 Page 40 of 610 REJ09B0131-0600
Section 2 CPU
* Arithmetic Operation Instructions
Instruction ADD ADD ADDC ADDV Rm,Rn #imm,Rn Rm,Rn Rm,Rn Operation
Rn + Rm Rn Rn + imm Rn Rn + Rm + T Rn, Carry T Rn + Rm Rn, Overflow T If R0 = imm, 1 T If Rn = Rm, 1 T If Rn Rm with unsigned data, 1 T If Rn Rm with signed data, 1 T If Rn > Rm with unsigned data, 1 T If Rn > Rm with signed data, 1 T If Rn 0, 1 T If Rn > 0, 1 T If Rn and Rm have an equivalent byte, 1 T Single-step division (Rn/Rm) MSB of Rn Q, MSB of Rm M, M^ Q T 0 M/Q/T
Code
0011nnnnmmmm1100 0111nnnniiiiiiii 0011nnnnmmmm1110
Execution Cycles T Bit 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 to 5*
Carry Overflow Comparison result Comparison result Comparison result Comparison result Comparison result Comparison result Comparison result Comparison result Comparison result Calculation result Calculation result
0011nnnnmmmm1111
CMP/EQ #imm,R0 CMP/EQ Rm,Rn CMP/HS Rm,Rn CMP/GE Rm,Rn CMP/HI Rm,Rn CMP/GT Rm,Rn CMP/PZ Rn CMP/PL Rn CMP/STRRm,Rn DIV1 DIV0S DIV0U DMULS.L Rm,Rn Rm,Rn
10001000iiiiiiii
0011nnnnmmmm0000
0011nnnnmmmm0010
0011nnnnmmmm0011
0011nnnnmmmm0110
0011nnnnmmmm0111
0100nnnn00010001
0100nnnn00010101
0010nnnnmmmm1100
0011nnnnmmmm0100
0010nnnnmmmm0111
0000000000011001 0011nnnnmmmm1101
0
Rm,Rn Signed operation of
Rn x Rm MACH, MACL 32 x 32 64 bits
Rev. 6.00 Jun. 12, 2007 Page 41 of 610 REJ09B0131-0600
Section 2 CPU
Instruction DMULU.L
Operation Rm,Rn Unsigned operation of
Rn x Rm MACH, MACL 32 x 32 64 bits
Code
0011nnnnmmmm0101
Execution Cycles T Bit 2 to 5*
DT
Rn
Rn - 1 Rn, if Rn = 0, 1 T, else 0 T A byte in Rm is signextended Rn A word in Rm is signextended Rn A byte in Rm is zeroextended Rn A word in Rm is zeroextended Rn Signed operation of (Rn) x (Rm) + MAC MAC, 32 x 32 + 64 64 bits Signed operation of (Rn) x (Rm) + MAC MAC, 16 x 16 + 64 64 bits Rn x Rm MACL 32 x 32 32 bits Signed operation of Rn x Rm MAC 16 x 16 32 bits Unsigned operation of Rn x Rm MAC 16 x 16 32 bits 0-Rm Rn 0-Rm-T Rn, Borrow T Rn-Rm Rn
0100nnnn00010000
1 1 1 1 1 2 to 5*
Comparison result
EXTS.B Rm,Rn EXTS.W Rm,Rn EXTU.B Rm,Rn EXTU.W Rm,Rn MAC.L @Rm+,@Rn+
0110nnnnmmmm1110

0110nnnnmmmm1111
0110nnnnmmmm1100
0110nnnnmmmm1101
0000nnnnmmmm1111
MAC.W @Rm+,@Rn+
0100nnnnmmmm1111
2 to 4*
MUL.L Rm,Rn MULS.W Rm,Rn
0000nnnnmmmm0111
2 to 5* 1 (3)*

0010nnnnmmmm1111
MULU.W Rm,Rn
0010nnnnmmmm1110
1 (3)*
NEG
Rm,Rn
0110nnnnmmmm1011 0110nnnnmmmm1010
1 1 1
Borrow
NEGC Rm,Rn
SUB
Rm,Rn
0011nnnnmmmm1000
Rev. 6.00 Jun. 12, 2007 Page 42 of 610 REJ09B0131-0600
Section 2 CPU
Instruction
Operation
Rn-Rm-T Rn, Borrow T Rn-Rm Rn, Underflow T
Code
0011nnnnmmmm1010
Execution Cycles 1 1
T Bit Borrow Overflow
SUBC Rm,Rn SUBV Rm,Rn
Note: *
0011nnnnmmmm1011
Indicates the number of execution cycles for normal operation. The values in parentheses indicate the number of execution cycles when conflicts occur with the previous or next instruction.
* Logic Operation Instructions
Instruction AND AND Rm,Rn #imm,R0 Operation
Rn & Rm Rn R0 & imm R0 (R0 + GBR)
Code
Execution Cycles
T Bit
Test result Test result Test result Test result
0010nnnnmmmm1001 1 11001001iiiiiiii 1 11001101iiiiiiii 3 0110nnnnmmmm0111 1 0010nnnnmmmm1011 1 11001011iiiiiiii 1 11001111iiiiiiii 3 0100nnnn00011011 4 0010nnnnmmmm1000 1 11001000iiiiiiii 1 11001100iiiiiiii 3 0010nnnnmmmm1010 1 11001010iiiiiiii 1 11001110iiiiiiii 3
AND.B #imm,@(R0,GBR) (R0 + GBR) & imm NOT OR OR Rm,Rn Rm,Rn #imm,R0
~Rm Rn Rn | Rm Rn R0 | imm R0 (R0 + GBR)
OR.B #imm,@(R0,GBR) (R0 + GBR) | imm TAS.B @Rn TST TST Rm,Rn #imm,R0
If (Rn) is 0, 1 T; 1 MSB of (Rn) Rn & Rm; if the result is 0, 1 T R0 & imm; if the result is 0, 1 T if the result is 0, 1 T
TST.B #imm,@(R0,GBR) (R0 + GBR) & imm; XOR XOR Rm,Rn #imm,R0
Rn ^ Rm Rn R0 ^ imm R0 (R0 + GBR)

XOR.B #imm,@(R0,GBR) (R0 + GBR) ^ imm
Rev. 6.00 Jun. 12, 2007 Page 43 of 610 REJ09B0131-0600
Section 2 CPU
* Shift Instructions
Instruction ROTL ROTR ROTCL ROTCR SHAL SHAR SHLL SHLR SHLL2 SHLR2 SHLL8 SHLR8 Rn Rn Rn Rn Rn Rn Rn Rn Rn Rn Rn Rn Operation
T Rn MSB LSB Rn T T Rn T T Rn T T Rn 0 MSB Rn T T Rn 0 0 Rn T Rn << 2 Rn Rn >> 2 Rn Rn << 8 Rn Rn >> 8 Rn Rn << 16 Rn Rn >> 16 Rn
Code
0100nnnn00000100 0100nnnn00000101 0100nnnn00100100 0100nnnn00100101 0100nnnn00100000 0100nnnn00100001 0100nnnn00000000 0100nnnn00000001 0100nnnn00001000 0100nnnn00001001 0100nnnn00011000 0100nnnn00011001 0100nnnn00101000 0100nnnn00101001
Execution Cycles 1 1 1 1 1 1 1 1 1 1 1 1 1 1
T Bit MSB LSB MSB LSB MSB LSB MSB LSB
SHLL16 Rn SHLR16 Rn
* Branch Instructions
Instruction BF label Operation
If T = 0, disp x 2 + PC PC; if T = 1, nop Delayed branch, if T = 0, disp x 2 + PC PC; if T = 1, nop If T = 1, disp x 2 + PC PC; if T = 0, nop Delayed branch, if T = 1, disp x 2 + PC PC; if T = 0, nop Delayed branch, disp x 2 + PC PC
Code
10001011dddddddd
Execution Cycles 3/1*
T Bit
BF/S label
10001111dddddddd
2/1*
BT
label
10001001dddddddd
3/1*
BT/S label
10001101dddddddd
2/1*
BRA
label
1010dddddddddddd
2
Rev. 6.00 Jun. 12, 2007 Page 44 of 610 REJ09B0131-0600
Section 2 CPU
Instruction BRAF Rm BSR label
Operation
Delayed branch, Rm + PC PC Delayed branch, PC PR, disp x 2 + PC PC Delayed branch, PC PR, Rm + PC PC Delayed branch, Rm PC Delayed branch, PC PR, Rm PC Delayed branch, PR PC
Code
0000mmmm00100011
Execution Cycles 2 2 2 2 2 2
T Bit
1011dddddddddddd
BSRF Rm JMP JSR RTS Note: * @Rm @Rm
0000mmmm00000011
0100mmmm00101011 0100mmmm00001011
0000000000001011
One cycle when the branch is not executed.
* System Control Instructions
Instruction CLRT CLRMAC LDC LDC LDC Rm,SR Rm,GBR Rm,VBR Operation 0T 0 MACH, MACL Rm SR Rm GBR Rm VBR
(Rm) GBR, Rm + 4 Rm (Rm) VBR, Rm + 4 Rm Rm MACH Rm MACL Rm PR (Rm) MACH, Rm + 4 Rm (Rm) MACL, Rm + 4 Rm
Code
0000000000001000 0000000000101000 0100mmmm00001110 0100mmmm00011110 0100mmmm00101110
Execution Cycles 1 1 6 4 4 8 4 4 1 1 1 1 1 1
T Bit 0 LSB LSB
LDC.L @Rm+,SR LDC.L @Rm+,GBR LDC.L @Rm+,VBR LDS LDS LDS Rm,MACH Rm,MACL Rm,PR
(Rm) SR, Rm + 4 Rm 0100mmmm00000111 0100mmmm00010111
0100mmmm00100111
0100mmmm00001010 0100mmmm00011010 0100mmmm00101010 0100mmmm00000110
LDS.L @Rm+,MACH LDS.L @Rm+,MACL LDS.L @Rm+,PR
0100mmmm00010110
(Rm) PR, Rm + 4 Rm 0100mmmm00100110
Rev. 6.00 Jun. 12, 2007 Page 45 of 610 REJ09B0131-0600
Section 2 CPU
Instruction NOP RTE SETT SLEEP STC STC STC SR,Rn GBR,Rn VBR,Rn
Operation
No operation Delayed branch, Stack area PC/SR 1T Sleep SR Rn GBR Rn VBR Rn Rn-4 Rn, SR (Rn) Rn-4 Rn, GBR (Rn) Rn-4 Rn, VBR (Rn) MACH Rn MACL Rn PR Rn
Code
0000000000001001 0000000000101011
Execution Cycles 1 5 1 4* 1 1 1 1 1 1 1 1 1 1 1 1 8
T Bit 1
0000000000011000 0000000000011011 0000nnnn00000010 0000nnnn00010010 0000nnnn00100010 0100nnnn00000011 0100nnnn00010011 0100nnnn00100011 0000nnnn00001010 0000nnnn00011010 0000nnnn00101010
STC.L SR,@-Rn STC.L GBR,@-Rn STC.L VBR,@-Rn STS STS STS MACH,Rn MACL,Rn PR,Rn
STS.L MACH,@-Rn STS.L MACL,@-Rn STS.L PR,@-Rn TRAPA #imm Note: *
Rn-4 Rn, MACH (Rn) 0100nnnn00000010 Rn-4 Rn, MACL (Rn) 0100nnnn00010010 Rn-4 Rn, PR (Rn) PC/SR Stack area, (imm x 4 + VBR) PC 0100nnnn00100010 11000011iiiiiiii
Number of execution cycles until this LSI enters sleep mode. About the number of execution cycles: The table lists the minimum number of execution cycles. In practice, the number of execution cycles will be increased depending on the conditions such as: * When there is a conflict between instruction fetch and data access * When the destination register of a load instruction (memory register) is also used by the instruction immediately after the load instruction.
Rev. 6.00 Jun. 12, 2007 Page 46 of 610 REJ09B0131-0600
Section 2 CPU
2.6
2.6.1
Processing States
State Transition
The CPU has the four processing states: reset, exception handling, program execution, and powerdown. Figure 2.4 shows the CPU state transition. Note that some products do not support the manual reset function and the MRES pin.
RES = 0 in any state RES = 1 and MRES = 0 in any state
Power-on reset state
Manual reset state
Reset state
Exception handling state Request for internal power-on reset Request for NMI or internal manual reset by the WDT or IRQ interrupt Request for End of exception handling exception handling Program execution state SLEEP instruction by clearing SSBY bit SLEEP instruction by setting SSBY bit
Sleep mode
Software standby mode
Power-down mode
Figure 2.4 CPU State Transition
Rev. 6.00 Jun. 12, 2007 Page 47 of 610 REJ09B0131-0600
Section 2 CPU
* Reset state The CPU is reset. When the RES pin is driven low, the CPU enters the power-on reset state. When the RES pin is high and MRES pin is low, the CPU enters the manual reset state. * Exception handling state This state is a transitional state in which the CPU processing state changes due to a request for exception handling such as a reset or an interrupt. When a reset occurs, the execution start address as the initial value of the program counter (PC) and the initial value of the stack pointer (SP) are fetched from the exception handling vector table. Then, a branch is made for the start address to execute a program. When an interrupt occurs, the PC and status register (SR) are saved in the stack area pointed to by SP. The start address of an exception handling routine is fetched from the exception handling vector table and a branch to the address is made to execute a program. Then the processing state enters the program execution state. * Program execution state The CPU executes programs sequentially. * Power-down state The CPU stops to reduce power consumption. The SLEEP instruction makes the CPU enter sleep mode or software standby mode.
Rev. 6.00 Jun. 12, 2007 Page 48 of 610 REJ09B0131-0600
Section 3 Cache
Section 3 Cache
3.1
* * * * * *
Features
Capacity: 4 kbytes (SH7618), 16 kbytes (SH7618A) Structure: Instructions/data unified, 4-way set associative Line size: 16 bytes Number of entries: 64 entries/way (SH7618), 256 entries/way (SH7618A) Write method: Write-back/write-through is selectable Replacement method: Least-recently-used (LRU) algorithm Cache Structure
3.1.1
The cache holds both instructions and data and employs a 4-way set associative system. It is composed of four ways (banks), and each of which is divided into an address section and a data section. Each of the address and data sections is divided into 64 entries (256 entries for the SH7618A). The data of an entry is called a line. Each line consists of 16 bytes (4 bytes x 4). The data capacity per way is 1 kbyte (16 bytes x 64 entries) (4 kbytes (16 bytes x 256 entries) for the SH7618A), with a total of 4 kbytes (16 kbytes for the SH7618A) in the cache (4 ways). Figure 3.1 shows the cache structure.
Address array (ways 0 to 3) Data array (ways 0 to 3) LRU
Entry 0 V U Tag address Entry 1 . . . . . .
0 1 . . . . . .
LW0
LW1
LW2
LW3
0 1 . . . . . .
Entry 63 (Entry 255)* 24 (1 + 1 + 22) bits
63 (255)* 128 (32 x 4) bits LW0 to LW3: Longword data 0 to 3
63 (255)* 6 bits
Note: * For the SH7618A.
Figure 3.1 Cache Structure
CACH000C_000020030900
Rev. 6.00 Jun. 12, 2007 Page 49 of 610 REJ09B0131-0600
Section 3 Cache
Address Array: The V bit indicates whether or not the entry data is valid. When the V bit is 1, data is valid; when 0, data is not valid. The U bit indicates whether or not the entry has been written to in write-back mode. When the U bit is 1, the entry has been written to; when 0, it has not. The tag address is composed of 22 bits (address bits 31 to 10) used for comparison during cache searches. In this LSI, the upper three bits of 32 address bits are used as shadow bits (see section 7, Bus State Controller (BSC)), therefore, the upper three bits of the tag address are cleared to 0. The V and U bits are initialized to 0 by a power-on reset. The tag address is not initialized by a power-on reset. Data Array: Holds 16-byte instruction and data. Entries are registered in the cache in line units (16 bytes). The data array is not initialized by a power-on reset. LRU: With the 4-way set associative system, up to four instructions or data with the same entry address can be registered in the cache. When an entry is registered, LRU shows which of the four ways it is registered in. There are six LRU bits, controlled by hardware. The least-recently-used (LRU) algorithm is used to select the way. When a cache miss occurs, six LRU bits indicate the way to be replaced. If a bit pattern other than those listed in table 3.1 is set in the LRU bits by software, the cache will not function correctly. When changing the LRU bits by software, set one of the patterns listed in table 3.1. The LRU bits are initialized to 000000 by a power-on reset. Table 3.1 LRU and Way to be Replaced
Way to be Replaced 3 2 1 0
LRU (Bits 5 to 0) 000000, 000100, 010100, 100000, 110000, 110100 000001, 000011, 001011, 100001, 101001, 101011 000110, 000111, 001111, 010110, 011110, 011111 111000, 111001, 111011, 111100, 111110, 111111
Rev. 6.00 Jun. 12, 2007 Page 50 of 610 REJ09B0131-0600
Section 3 Cache
3.1.2
Divided Areas and Cache
A 4-G byte address space is divided into five areas with the architecture of this LSI. The cache access methods can be specified for each area. Table 3.2 lists the correspondence between the divided areas and cache. Table 3.2
Address H'00000000 to H'7FFFFFFF H'80000000 to H'9FFFFFFF H'A0000000 to H'BFFFFFFF H'C0000000 to H'DFFFFFFF H'E0000000 to H'FFFFFFFF
Correspondence between Divided Areas and Cache
Area P0 P1 P2 P3 P4 Cacheable Cacheable Cacheable Non cacheable Cacheable Cache Operating Control WT bit in CCR1 CB bit in CCR1 WT bit in CCR1
Non cacheable (internal I/O)
Rev. 6.00 Jun. 12, 2007 Page 51 of 610 REJ09B0131-0600
Section 3 Cache
3.2
Register Descriptions
The cache has the following registers. For details on register addresses and register states during each process, refer to section 20, List of Registers. * Cache control register 1 (CCR1) * Cache control register 3 (CCR3)* Note: * Supported only by the SH7618. 3.2.1 Cache Control Register 1 (CCR1)
The cache is enabled or disabled by the CE bit in CCR1. CCR1 also has the CF bit (which invalidates all cache entries), and the WT and CB bits (which select either write-through mode or write-back mode). Programs that change the contents of CCR1 should be placed in the address space that is not cached.
Bit 31 to 4 Bit Name -- Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 3 CF 0 R/W Cache Flush Writing 1 flushes all cache entries meaning that it clears the V, U, and LRU bits of all cache entries to 0. This bit is always read as 0. Write-back to external memory is not performed when the cache is flushed. 2 CB 0 R/W Write-Back Indicates the cache operating mode for H'80000000 to H'9FFFFFFF. 0: Write-through mode 1: Write-back mode 1 WT 0 R/W Write-Through Indicates the cache operating mode for H'00000000 to H'7FFFFFFF and H'C0000000 to H'DFFFFFFF. 0: Write-back mode 1: Write-through mode
Rev. 6.00 Jun. 12, 2007 Page 52 of 610 REJ09B0131-0600
Section 3 Cache
Bit 0
Bit Name CE
Initial Value 0
R/W R/W
Description Cache Enable Indicates whether or not the cache function is used. 0: Cache function is not used. 1: Cache function is used.
3.2.2
Cache Control Register 3 (CCR3)
CCR3 specifies the cache size. Programs that change the contents of CCR3 should be placed in the address space that is not cached. Note: Supported only by the SH7618.
Bit 31 to 17 Bit Name -- Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 16 15 14 13 to 0 CSIZE2 CSIZE1 CSIZE0 -- 0 0 1 All 0 R/W R/W R/W R Cache Size Writing B'100 to these bits specifies the cache size 16 kbytes. Write B'100 before enabling the cache by the CE bit in CCR1. Reserved These bits are always read as 0. The write value should always be 0.
Rev. 6.00 Jun. 12, 2007 Page 53 of 610 REJ09B0131-0600
Section 3 Cache
3.3
3.3.1
Operation
Searching Cache
If the cache is enabled (the CE bit in CCR1 is set to 1), whenever an instruction or data in H'00000000 to H'7FFFFFFF, H'8000000 to H'9FFFFFFF, and H'C0000000 to H'DFFFFFFF is accessed, the cache will be searched to see if the desired instruction or data is in the cache. Figure 3.2 illustrates the method by which the cache is searched. Entries are selected using bits 9 to 4 (bits 11 to 4 for the SH7618A) of the memory access address and the tag address of that entry is read. The address comparison is performed on all four ways. When the comparison shows a match and the selected entry is valid (V = 1), a cache hit occurs. When the comparison does not show a match or the selected entry is not valid (V = 0), a cache miss occurs. Figure 3.2 shows a hit on way 1.
Rev. 6.00 Jun. 12, 2007 Page 54 of 610 REJ09B0131-0600
Section 3 Cache
Address
31 10 (12*) 9 (11*) 4 3 2 10
Entry selection
Longword (LW) selection Ways 0 to 3 Ways 0 to 3
0 1
V U Tag address
LW0
LW1
LW2
LW3
63 (255*)
CMP0 CMP1 CMP2 CMP3
Hit signal 1 CMP0: Comparison circuit 0 CMP1: Comparison circuit 1 CMP2: Comparison circuit 2 CMP3: Comparison circuit 3 Note: * For the SH7618A.
Figure 3.2 Cache Search Scheme
Rev. 6.00 Jun. 12, 2007 Page 55 of 610 REJ09B0131-0600
Section 3 Cache
3.3.2
Read Access
Read Hit: In a read access, instructions and data are transferred from the cache to the CPU. The LRU bits are updated so that they point to the most recently hit way. Read Miss: An external bus cycle starts and the entry is updated. The way to be replaced is shown in table 3.1. Data is updated in units of 16 bytes by updating the entry. When the desired instruction or data is loaded from external memory to the cache, the instruction or data is transferred to the CPU in parallel. When it is loaded to the cache, the U bit is cleared to 0, the V bit is set to 1, the LRU bits are updated so that they point to the most recently hit way. When the U bit of the entry which is to be replaced by entry updating in write-back mode is 1, the cacheupdate cycle starts after the entry is transferred to the write-back buffer. After the cache completes its update cycle, the write-back buffer writes the entry back to the memory. Transfer is in 16-byte units. 3.3.3 Write Access
Write Hit: In a write access in write-back mode, the data is written to the cache and no external memory write cycle is generated. The U bit of the entry that has been written to is set to 1, and the LRU bits are updated to indicate that the hit way is the most recently hit way. In write-through mode, the data is written to the cache and an external memory write cycle is generated. The U bit of the entry that has been written to is not updated, and the LRU bits are updated to indicate that the hit way is the most recently hit way. Write Miss: In write-back mode, an external write cycle starts when a write miss occurs, and the entry is updated. The way to be replaced is shown in table 3.1. When the U bit of the entry which is to be replaced by entry updating is 1, the cache-update cycle starts after the entry has been transferred to the write-back buffer. Data is written to the cache and the U bit and the V bit are set to 1. The LRU bits are updated to indicate that the replaced way is the most recently updated way. After the cache has completed its update cycle, the write-back buffer writes the entry back to the memory. Transfer is in 16-byte units. In write-through mode, no write to cache occurs in a write miss; the write is only to the external memory.
Rev. 6.00 Jun. 12, 2007 Page 56 of 610 REJ09B0131-0600
Section 3 Cache
3.3.4
Write-Back Buffer
When the U bit of the entry to be replaced in write-back mode is 1, the entry must be written back to the external memory. To increase performance, the entry to be replaced is first transferred to the write-back buffer and fetching of new entries to the cache takes priority over writing back to the external memory. After the fetching of new entries to the cache completes, the write-back buffer writes the entry back to the external memory. During the write-back cycles, the cache can be accessed. The write-back buffer can hold one line of cache data (16 bytes) and its physical address. Figure 3.3 shows the configuration of the write-back buffer.
PA (31 to 4) Longword 0 Longword 1 Longword 2 Longword 3 PA (31 to 4): Physical address to be written to external memory Longword 0 to 3: One line of cache data to be written to external memory
Figure 3.3 Write-Back Buffer Configuration 3.3.5 Coherency of Cache and External Memory
Coherency between the cache and the external memory must be ensured by software. When memory shared by this LSI and another device is allocated to a cacheable address space, invalidate and write back the cache by accessing the memory-mapped cache, as required. Memory that is shared by the CPU and E-DMAC of this LSI should also be handled in this way.
Rev. 6.00 Jun. 12, 2007 Page 57 of 610 REJ09B0131-0600
Section 3 Cache
3.4
Memory-Mapped Cache
To allow software management of the cache, cache contents can be read from or written to by the MOV instructions. The address array is allocated to addresses H'F0000000 to H'F0FFFFFF, and the data array to addresses H'F1000000 to H'F1FFFFFF. The address array and data array must be accessed in longwords, and instruction fetches cannot be performed. 3.4.1 Address Array
The address array is allocated to H'F0000000 to H'F0FFFFFF. To access an address array, the 32bit address field (for read/write accesses) and 32-bit data field (for write accesses) must be specified. The address field specifies information for selecting the entry to be accessed; the data field specifies the tag address, V bit, U bit, and LRU bits to be written to the address array. In the address field, specify the entry address for selecting the entry, W for selecting the way, A for enabling or disabling the associative operation, and H'F0 for indicating address array access. As for W, 00 indicates way 0, 01 indicates way 1, 10 indicates way 2, and 11 indicates way 3. In the data field, specify the tag address, LRU bits, U bit, and V bit. Always clear the upper three bits (bits 31 to 29) of the tag address to 0. Figure 3.4 shows the address and data formats. The following three operations are available in the address array. Address-Array Read: Read the tag address, LRU bits, U bit, and V bit for the entry that corresponds to the entry address and way specified by the address field of the read instruction. In reading, the associative operation is not performed, regardless of whether the associative bit (A bit) specified in the address is 1 or 0. Address-Array Write (Non-Associative Operation): Write the tag address, LRU bits, U bit, and V bit, specified by the data field of the write instruction, to the entry that corresponds to the entry address and way as specified by the address field of the write instruction. Ensure that the associative bit (A bit) in the address field is set to 0. When writing to a cache line for which the U bit = 1 and the V bit =1, write the contents of the cache line back to memory, then write the tag address, LRU bits, U bit, and V bit specified by the data field of the write instruction. When 0 is written to the V bit, 0 must also be written to the U bit for that entry. Address-Array Write (Associative Operation): When writing with the associative bit (A bit) of the address field set to 1, the addresses in the four ways for the entry specified by the address field of the write instruction are compared with the tag address that is specified by the data field of the write instruction. Write the U bit and the V bit specified by the data field of the write instruction to the entry of the way that has a hit. However, the tag address and LRU bits remain unchanged. When there is no way that has a hit, nothing is written and there is no operation. This function is
Rev. 6.00 Jun. 12, 2007 Page 58 of 610 REJ09B0131-0600
Section 3 Cache
used to invalidate a specific entry in the cache. When the U bit of the entry that has had a hit is 1 at this time, writing back should be performed. However, when 0 is written to the V bit, 0 must also be written to the U bit of that entry. 3.4.2 Data Array
The data array is allocated to H'F1000000 to H'F1FFFFFF. To access a data array, the 32-bit address field (for read/write accesses) and 32-bit data field (for write accesses) must be specified. The address field specifies information for selecting the entry to be accessed; the data field specifies the longword data to be written to the data array. In the address field, specify the entry address for selecting the entry, L for indicating the longword position within the (16-byte) line, W for selecting the way, and H'F1 for indicating data array access. As for L, 00 indicates longword 0, 01 indicates longword 1, 10 indicates longword 2, and 11 indicates longword 3. As for W, 00 indicates way 0, 01 indicates way 1, 10 indicates way 2, and 11 indicates way 3. Since access size of the data array is fixed at longword, bits 1 and 0 of the address field should be set to 00. Figure 3.4 shows the address and data formats. The following two operations on the data array are available. The information in the address array is not affected by these operations. Data-Array Read: Read the data specified by L of the address field, from the entry that corresponds to the entry address and the way that is specified by the address field. Data-Array Write: Write the longword data specified by the data field, to the position specified by L of the address field, in the entry that corresponds to the entry address and the way specified by the address field.
Rev. 6.00 Jun. 12, 2007 Page 59 of 610 REJ09B0131-0600
Section 3 Cache
(1) Address array access (a) Address specification Read access 31 1111 0000 Write access 31 1111 0000 (14*)(13*)(12*)(11*) 12 11 10 9 *--------* W Entry address
24 23
43210 0*00
24 23 *--------*
(14*)(13*)(12*)(11*) 12 11 10 9 W Entry address
43210 A*00
(b) Data specification (both read and write accesses) 31 30 29 28 00 0 Tag address (28 to 10) 10 9 LRU 43210 X XU V
(2) Data array access (both read and write accesses) (a) Address specification 31 1111 0001 (b) Data specification 31 Longword [Legend] *: Don't care X: 0 for read, don't care for write Note: * For the SH7618A. 0 24 23 *--------* (14*)(13*)(12*)(11*) 12 11 10 9 W Entry address 43210 L 00
Figure 3.4 Specifying Address and Data for Memory-Mapped Cache Access
Rev. 6.00 Jun. 12, 2007 Page 60 of 610 REJ09B0131-0600
Section 3 Cache
3.4.3
Usage Examples
Invalidating Specific Entries: Specific cache entries can be invalidated by writing 0 to the entry's V bit in the memory-mapped cache access. When the A bit is 1, the tag address specified by the write data is compared to the tag address within the cache selected by the entry address, and the V bit and U bit specified by the write data are written when a match is found. If no match is found, there is no operation. When the V bit of an entry in the address array is set to 0, the entry is written back if the entry's U bit is 1. In the example shown below, R0 specifies the write data and R1 specifies the address.
; R0=H'01100010; VPN=B'0000 0001 0001 0000 0000 00, U=0, V=0 ; R1=H'F0000088; address array access, entry=B'001000 (entry=B'00001000 for the SH7618A), A=1 ; MOV.L R0,@R1
Reading Data of Specific Entry: The data section of a specific entry can be read from by the memory-mapped cache access. The longword indicated in the data field of the data array in figure 3.4 is read into the register. In the example shown below, R0 specifies the address and R1 shows what is read.
; R0=H'F100004C; data array access, entry=B'000100 (entry=B'00000100 for the SH7618A) ; Way = 0, longword address = 3 ; MOV.L @R0,R1 ; Longword 3 is read.
Rev. 6.00 Jun. 12, 2007 Page 61 of 610 REJ09B0131-0600
Section 3 Cache
Rev. 6.00 Jun. 12, 2007 Page 62 of 610 REJ09B0131-0600
Section 4 U Memory
Section 4 U Memory
This LSI has on-chip U memory which can be used to store instructions and data.
4.1
Features
Features of the U Memory are shown below. * Size 4 kbytes * Address H'E55FF000 to H'E55FFFFF * Priority The U memory can be accessed from the I bus by the E-DMAC and from the L bus by the CPU. In the event of simultaneous accesses from different buses, the accesses are processed according to the priority. The priority is: I bus > L bus.
4.2
Usage Notes
In sleep mode, the U memory cannot be accessed by the E-DMAC.
RAM04K0A_000020030900
Rev. 6.00 Jun. 12, 2007 Page 63 of 610 REJ09B0131-0600
Section 4 U Memory
Rev. 6.00 Jun. 12, 2007 Page 64 of 610 REJ09B0131-0600
Section 5 Exception Handling
Section 5 Exception Handling
5.1
5.1.1
Overview
Types of Exception Handling and Priority
Exception handling is started by four sources: resets, address errors, interrupts and instructions and have the priority, as shown in table 5.1. When several exceptions are detected at once, they are processed according to the priority. Table 5.1
Exception Reset
Types of Exceptions and Priority
Exception Source Power-on reset H-UDI reset Priority High
Interrupt Address error Instruction
User break (break before instruction execution) CPU address error (instruction fetch) General illegal instructions (undefined code) Illegal slot instruction (undefined code placed immediately after a delayed branch instruction*1 or instruction that changes the PC value*2) Trap instruction (TRAPA instruction)
Address error Interrupt
CPU address error (data access) User break (break after instruction execution or operand break) NMI H-UDI IRQ On-chip peripheral modules: Watchdog timer (WDT) Ether controller (EtherC and E-DMAC) Compare match timer 0 and 1 (CMT0 and CMT1) Serial communication interface with FIFO (SCIF0, SCIF1, and SCIF2) Host interface (HIF) Low
Notes: 1. Delayed branch instructions: JMP, JSR, BRA, BSR, RTS, RTE, BF/S, BT/S, BSRF, and BRAF. 2. Instructions that change the PC value: JMP, JSR, BRA, BSR, RTS, RTE, BT, BF, TRAPA, BF/S, BT/S, BSRF, BRAF, LDC Rm,SR, LDC.L @Rm+,SR.
Rev. 6.00 Jun. 12, 2007 Page 65 of 610 REJ09B0131-0600
Section 5 Exception Handling
5.1.2
Exception Handling Operations
The exceptions are detected and the exception handling starts according to the timing shown in table 5.2. Table 5.2
Exception Reset Power-on reset H-UDI reset Address error Interrupt Instruction Trap instruction General illegal instructions Illegal slot instructions
Timing for Exception Detection and Start of Exception Handling
Timing of Source Detection and Start of Exception Handling Started when the RES pin changes from low to high or when the WDT overflows. Started when the reset assert command and the reset negate command are input to the H-UDI in this order. Detected during the instruction decode stage and started after the execution of the current instruction is completed. Started by the execution of the TRAPA instruction. Started when an undefined code placed at other than a delay slot (immediately after a delayed branch instruction) is decoded. Started when an undefined code placed at a delay slot (immediately after a delayed branch instruction) or an instruction that changes the PC value is detected.
When exception handling starts, the CPU operates Exception Handling Triggered by Reset: The initial values of the program counter (PC) and stack pointer (SP) are fetched from the exception handling vector table (PC from the address H'A0000000 and SP from the address H'A0000004). For details, see section 5.1.3, Exception Handling Vector Table. H'00000000 is then written to the vector base register (VBR), and H'F (B'1111) is written to the interrupt mask bits (I3 to I0) in the status register (SR). The program starts from the PC address fetched from the exception handling vector table. Exception Handling Triggered by Address Error, Interrupt, and Instruction: SR and PC are saved to the stack indicated by R15. For interrupt exception handling, the interrupt priority level is written to the interrupt mask bits (I3 to I0) in SR. For address error and instruction exception handling, bits I3 to I0 are not affected. The start address is then fetched from the exception handling vector table and the program starts from that address.
Rev. 6.00 Jun. 12, 2007 Page 66 of 610 REJ09B0131-0600
Section 5 Exception Handling
5.1.3
Exception Handling Vector Table
Before exception handling starts, the exception handling vector table must be set in memory. The exception handling vector table stores the start addresses of exception handling routines. (The reset exception handling table holds the initial values of PC and SP.) All exception sources are given different vector numbers and vector table address offsets. The vector table addresses are calculated from these vector numbers and vector table address offsets. During exception handling, the start addresses of the exception handling routines are fetched from the exception handling vector table that is indicated by this vector table address. Table 5.3 shows the vector numbers and vector table address offsets. Table 5.4 shows how vector table addresses are calculated. Table 5.3 Vector Numbers and Vector Table Address Offsets
Vector Number 0 1 2 3 General illegal instruction (Reserved by system) Illegal slot instruction (Reserved by system) 4 5 6 7 8 CPU address error (Reserved by system) Interrupt NMI User break H-UDI (Reserved by system) 9 10 11 12 13 14 : 31 Vector Table Address Offset H'00000000 to H'00000003 H'00000004 to H'00000007 H'00000008 to H'0000000B H'0000000C to H'0000000F H'00000010 to H'00000013 H'00000014 to H'00000017 H'00000018 to H'0000001B H'0000001C to H'0000001F H'00000020 to H'00000023 H'00000024 to H'00000027 H'00000028 to H'0000002B H'0000002C to H'0000002F H'00000030 to H'00000033 H'00000034 to H'00000037 H'00000038 to H'0000003B : H'0000007C to H'0000007F
Exception Handling Source Power-on reset H-UDI reset (Reserved by system) PC SP
Rev. 6.00 Jun. 12, 2007 Page 67 of 610 REJ09B0131-0600
Section 5 Exception Handling
Exception Handling Source Trap instruction (user vector)
Vector Number 32 : 63
Vector Table Address Offset H'00000080 to H'00000083 : H'000000FC to H'000000FF H'00000100 to H'00000103 H'00000104 to H'00000107 H'00000108 to H'0000010B H'0000010C to H'0000010F H'00000110 to H'00000113 : H'0000013C to H'0000013F H'00000140 to H'00000143 H'00000144 to H'00000147 H'00000148 to H'0000014B H'0000014C to H'0000014F H'00000120 to H'00000124 : H'000003FC to H'000003FF
Interrupt
IRQ0 IRQ1 IRQ2 IRQ3 (Reserved by system)
64 65 66 67 68 : 79
IRQ4 IRQ5 IRQ6 IRQ7 On-chip peripheral module*
80 81 82 83 84 : 255
Note:
*
For details on the vector numbers and vector table address offsets of on-chip peripheral module interrupts, see table 6.2, Interrupt Exception Handling Vectors and Priorities in section 6, Interrupt Controller (INTC).
Table 5.4
Calculating Exception Handling Vector Table Addresses
Vector Table Address Calculation Vector table address = H'A0000000 + (vector table address offset) = H'A0000000 + (vector number) x 4 Vector table address = VBR + (vector table address offset) = VBR + (vector number) x 4
Exception Source Resets Address errors, interrupts, instructions
Notes: 1. VBR: Vector base register 2. Vector table address offset: See table 5.3. 3. Vector number: See table 5.3.
Rev. 6.00 Jun. 12, 2007 Page 68 of 610 REJ09B0131-0600
Section 5 Exception Handling
5.2
5.2.1
Resets
Types of Resets
Resets have priority over any exception source. As table 5.5 shows, a power-on reset initializes all modules in this LSI. Table 5.5 Reset Status
Conditions for Transition to Reset State WDT Overflow Overflow Not overflowed H-UDI Command Internal State On-Chip Peripheral Module Initialized Initialized Initialized
Type Power-on reset
RES Low High
CPU, INTC Initialized Initialized
PFC, I/O Port Initialized Initialized Initialized
H-UDI reset
High
Reset assert Initialized command
5.2.2
Power-On Reset
Power-On Reset by RES Pin: When the RES pin is driven low, this LSI enters the power-on reset state. To reliably reset this LSI, the RES pin should be kept low for at least the oscillation settling time when applying the power or when in standby mode (when the clock is halted) or at least 20 tcyc when the clock is operating. During the power-on reset state, CPU internal states and all registers of on-chip peripheral modules are initialized. In the power-on reset state, power-on reset exception handling starts when driving the RES pin high after driving the pin low for the given time. The CPU operates as follows: 1. The initial value (execution start address) of the program counter (PC) is fetched from the exception handling vector table. 2. The initial value of the stack pointer (SP) is fetched from the exception handling vector table. 3. The vector base register (VBR) is cleared to H'00000000 and the interrupt mask bits (I3 to I0) of the status register (SR) are set to H'F (B'1111). 4. The values fetched from the exception handling vector table are set in PC and SP, then the program starts. Be certain to always perform power-on reset exception handling when turning the system power on.
Rev. 6.00 Jun. 12, 2007 Page 69 of 610 REJ09B0131-0600
Section 5 Exception Handling
Power-On Reset by WDT: When TCNT of the WDT overflows while a setting is made so that a power-on reset can be generated in watchdog timer mode of the WDT, this LSI enters the poweron reset state. If a reset caused by the signal input on the RES pin and a reset caused by a WDT overflow occur simultaneously, the RES pin reset has priority, and the WOVF bit in RSTCSR is cleared to 0. When the power-on reset exception handling caused by the WDT is started, the CPU operates as follows: 1. The initial value (execution start address) of the program counter (PC) is fetched from the exception handling vector table. 2. The initial value of the stack pointer (SP) is fetched from the exception handling vector table. 3. The vector base register (VBR) is cleared to H'00000000 and the interrupt mask bits (I3 to I0) of the status register (SR) are set to H'F (B'1111). 4. The values fetched from the exception handling vector table are set in the PC and SP, then the program starts. 5.2.3 H-UDI Reset
The H-UDI reset is generated by issuing the H-UDI reset assert command. The CPU operation is described below. For details, see section 19, User Debugging Interface (H-UDI). 1. The initial value (execution start address) of the program counter (PC) is fetched from the exception handling vector table. 2. The initial value of the stack pointer (SP) is fetched from the exception handling vector table. 3. The vector base register (VBR) is cleared to H'00000000 and the interrupt mask bits (I3 to I0) in the status register (SR) are set to H'F (B'1111). 4. The values fetched from the exception handling vector table are set in PC and SP, then the program starts.
Rev. 6.00 Jun. 12, 2007 Page 70 of 610 REJ09B0131-0600
Section 5 Exception Handling
5.3
5.3.1
Address Errors
Address Error Sources
Address errors occur when instructions are fetched or data is read from or written to, as shown in table 5.6. Table 5.6 Bus Cycles and Address Errors
Bus Cycle Type Instruction fetch Data read/write Bus Master CPU Bus Cycle Description Instruction fetched from even address Instruction fetched from odd address CPU Word data accessed from even address Word data accessed from odd address Longword data accessed from a longword boundary Longword data accessed from other than a long-word boundary Address Errors None (normal) Address error occurs None (normal) Address error occurs None (normal) Address error occurs
5.3.2
Address Error Exception Source
When an address error exception is generated, the bus cycle which caused the address error ends, the current instruction finishes, and then the address error exception handling starts. The CPU operates as follows: 1. The status register (SR) is saved to the stack. 2. The program counter (PC) is saved to the stack. The PC value to be saved is the start address of the instruction which caused an address error exception. When the instruction that caused the exception is placed in the delay slot, the address of the delayed branch instruction which is placed immediately before the delay slot. 3. The start address of the exception handling routine is fetched from the exception handling vector table that corresponds to the generated address error, and the program starts executing from that address. This branch is not a delayed branch.
Rev. 6.00 Jun. 12, 2007 Page 71 of 610 REJ09B0131-0600
Section 5 Exception Handling
5.4
5.4.1
Interrupts
Interrupt Sources
Table 5.7 shows the sources that start the interrupt exception handling. They are NMI, user break, H-UDI, IRQ and on-chip peripheral modules. Table 5.7
Type NMI User break H-UDI IRQ On-chip peripheral module
Interrupt Sources
Request Source NMI pin (external input) User break controller (UBC) User debug interface (H-UDI) IRQ0 to IRQ7 pins (external input) Watchdog timer (WDT) Ether controller (EtherC and E-DMAC) Compare match timer (CMT0 and CMT1) FIFO on-chip serial communication interface (SCIF0, SCIF1, and SCIF2) Host interface (HIF) Number of Sources 1 1 1 8 1 1 2 12 2
All interrupt sources are given different vector numbers and vector table address offsets. For details on vector numbers and vector table address offsets, see table 6.2, Interrupt Exception Handling Vectors and Priorities in section 6, Interrupt Controller (INTC).
Rev. 6.00 Jun. 12, 2007 Page 72 of 610 REJ09B0131-0600
Section 5 Exception Handling
5.4.2
Interrupt Priority
The interrupt priority is predetermined. When multiple interrupts occur simultaneously (overlapped interruptions), the interrupt controller (INTC) determines their relative priorities and starts the exception handling according to the results. The priority of interrupts is expressed as priority levels 0 to 16, with priority 0 the lowest and priority 16 the highest. The NMI interrupt has priority 16 and cannot be masked, so it is always accepted. The priority level of the user break interrupt and H-UDI is 15. IRQ interrupt and on-chip peripheral module interrupt priority levels can be set freely using the interrupt priority level setting registers A to E (IPRA to IPRE) of the INTC as shown in table 5.8. The priority levels that can be set are 0 to 15. Level 16 cannot be set. For details on IPRA to IPRE, see section 6.3.4, Interrupt Priority Registers A to E (IPRA to IPRE). Table 5.8
Type NMI User break H-UDI IRQ On-chip peripheral module
Interrupt Priority
Priority Level 16 15 15 0 to 15 Comment Fixed priority level. Cannot be masked. Fixed priority level. Can be masked. Fixed priority level. Set with interrupt priority level setting registers A through E (IPRA to IPRE).
5.4.3
Interrupt Exception Handling
When an interrupt occurs, the interrupt controller (INTC) ascertains its priority level. NMI is always accepted, but other interrupts are only accepted if they have a priority level higher than the priority level set in the interrupt mask bits (I3 to I0) of the status register (SR). When an interrupt is accepted, exception handling begins. In interrupt exception handling, the CPU saves SR and the program counter (PC) to the stack. The priority level of the accepted interrupt is written to bits I3 to I0 in SR. Although the priority level of the NMI is 16, the value set in bits I3 to I0 is H'F (level 15). Next, the start address of the exception handling routine is fetched from the exception handling vector table for the accepted interrupt, and program execution branches to that address and the program starts. For details on the interrupt exception handling, see section 6.6, Interrupt Operation.
Rev. 6.00 Jun. 12, 2007 Page 73 of 610 REJ09B0131-0600
Section 5 Exception Handling
5.5
5.5.1
Exceptions Triggered by Instructions
Types of Exceptions Triggered by Instructions
Exception handling can be triggered by the trap instruction, illegal slot instructions, and general illegal instructions, as shown in table 5.9. Table 5.9
Type Trap instruction Illegal slot instructions*
Types of Exceptions Triggered by Instructions
Source Instruction TRAPA Undefined code placed immediately after a delayed branch instruction (delay slot) or instructions that changes the PC value Comment Delayed branch instructions: JMP, JSR, BRA, BSR, RTS, RTE, BF/S, BT/S, BSRF, BRAF Instructions that changes the PC value: JMP, JSR, BRA, BSR, RTS, RTE, BT, BF, TRAPA, BF/S, BT/S, BSRF, BRAF, LDC Rm,SR, LDC.L @Rm+,SR
General illegal instructions* Note: *
Undefined code anywhere besides in a delay slot
The operation is not guaranteed when undefined instructions other than H'FC00 to H'FFFF are decoded.
5.5.2
Trap Instructions
When a TRAPA instruction is executed, the trap instruction exception handling starts. The CPU operates as follows: 1. The status register (SR) is saved to the stack. 2. The program counter (PC) is saved to the stack. The PC value saved is the start address of the instruction to be executed after the TRAPA instruction. 3. The CPU reads the start address of the exception handling routine from the exception handling vector table that corresponds to the vector number specified in the TRAPA instruction, program execution branches to that address, and then the program starts. This branch is not a delayed branch.
Rev. 6.00 Jun. 12, 2007 Page 74 of 610 REJ09B0131-0600
Section 5 Exception Handling
5.5.3
Illegal Slot Instructions
An instruction placed immediately after a delayed branch instruction is called "instruction placed in a delay slot". When the instruction placed in the delay slot is an undefined code, illegal slot exception handling starts after the undefined code is decoded. Illegal slot exception handling also starts when an instruction that changes the program counter (PC) value is placed in a delay slot and the instruction is decoded. The CPU handles an illegal slot instruction as follows: 1. The status register (SR) is saved to the stack. 2. The program counter (PC) is saved to the stack. The PC value saved is the target address of the delayed branch instruction immediately before the undefined code or the instruction that rewrites the PC. 3. The start address of the exception handling routine is fetched from the exception handling vector table that corresponds to the exception that occurred. Program execution branches to that address and the program starts. This branch is not a delayed branch. 5.5.4 General Illegal Instructions
When an undefined code placed anywhere other than immediately after a delayed branch instruction (i.e., in a delay slot) is decoded, general illegal instruction exception handling starts. The CPU handles the general illegal instructions in the same procedures as in the illegal slot instructions. Unlike processing of illegal slot instructions, however, the program counter value that is stacked is the start address of the undefined code.
Rev. 6.00 Jun. 12, 2007 Page 75 of 610 REJ09B0131-0600
Section 5 Exception Handling
5.6
Cases when Exceptions are Accepted
When an exception other than resets occurs during decoding the instruction placed in a delay slot or immediately after an interrupt disabled instruction, it may not be accepted and be held shown in table 5.10. In this case, when an instruction which accepts an interrupt request is decoded, the exception is accepted. Table 5.10 Delay Slot Instructions, Interrupt Disabled Instructions, and Exceptions
Exception Address Error x*
2
Occurrence Timing Instruction in delay slot
General Illegal Instruction
Slot Illegal Instruction x*
2
Trap Instruction
Interrupt x*3 x*4
Immediately after interrupt disabled instruction*1
[Legend] : Accepted x: Not accepted : Does not occur Notes: 1. Interrupt disabled instructions: LDC, LDC.L, STC, STC.L, LDS, LDS.L, STS, and STS.L 2. An exception is accepted before the execution of a delayed branch instruction. However, when an address error or a slot illegal instruction exception occurs in the delay slot of the RTE instruction, correct operation is not guaranteed. 3. An exception is accepted after a delayed branch (between instructions in the delay slot and the branch destination). 4. An exception is accepted after the execution of the next instruction of an interrupt disabled instruction (before the execution two instructions after an interrupt disabled instruction).
Rev. 6.00 Jun. 12, 2007 Page 76 of 610 REJ09B0131-0600
Section 5 Exception Handling
5.7
Stack States after Exception Handling Ends
The stack states after exception handling ends are shown in table 5.11. Table 5.11 Stack Status after Exception Handling Ends
Types Address error (when the instruction that caused an exception is placed in the delay slot) Stack State
SP
Address of delayed branch instruction SR
32 bits
32 bits
Address error (other than above)
SP Address of instruction that caused exception SR 32 bits
32 bits
Interrupt
SP Address of instruction after executed instruction SR 32 bits
32 bits
Trap instruction
SP Address of instruction after TRAPA instruction SR 32 bits
32 bits
Rev. 6.00 Jun. 12, 2007 Page 77 of 610 REJ09B0131-0600
Section 5 Exception Handling
Types Illegal slot instruction
Stack State
SP
Address of delayed branch instruction SR
32 bits
32 bits
General illegal instruction
SP Address of general illegal instruction SR 32 bits
32 bits
Rev. 6.00 Jun. 12, 2007 Page 78 of 610 REJ09B0131-0600
Section 5 Exception Handling
5.8
5.8.1
Usage Notes
Value of Stack Pointer (SP)
The SP value must always be a multiple of 4. If it is not, an address error will occur when the stack is accessed during exception handling. 5.8.2 Value of Vector Base Register (VBR)
The VBR value must always be a multiple of 4. If it is not, an address error will occur when the stack is accessed during exception handling. 5.8.3 Address Errors Caused by Stacking for Address Error Exception Handling
When the SP value is not a multiple of 4, an address error will occur when stacking for exception handling (interrupts, etc.) and address error exception handling will start after the first exception handling is ended. Address errors will also occur in the stacking for this address error exception handling. To ensure that address error exception handling does not go into an endless loop, no address errors are accepted at that point. This allows program control to be passed to the handling routine for address error exception and enables error processing. When an address error occurs during exception handling stacking, the stacking bus cycle (write) is executed. When stacking the SR and PC values, the SP values for both are subtracted by 4, therefore, the SP value is still not a multiple of 4 after the stacking. The address value output during stacking is the SP value whose lower two bits are cleared to 0. So the write data stacked is undefined. 5.8.4 Notes on Slot Illegal Instruction Exception Handling
Some specifications on slot illegal instruction exception handling in this LSI differ from those on the conventional SH2. * * Conventional SH2: Instructions LDC Rm,SR and LDC.L @Rm+,SR are not subject to the slot illegal instructions. This LSI: Instructions LDC Rm,SR and LDC.L @Rm+,SR are subject to the slot illegal instructions.
The supporting status on our software products regarding this note is as follows:
Rev. 6.00 Jun. 12, 2007 Page 79 of 610 REJ09B0131-0600
Section 5 Exception Handling
Compiler This instruction is not allocated in the delay slot in the compiler V.4 or later versions. Real-time OS for ITRON specifications 1. HI7000/4, HI-SH7 This instruction does not exist in the delay slot within the OS. 2. HI7000 This instruction is in part allocated to the delay slot within the OS, which may cause the slot illegal instruction exception handling in this LSI. 3. Others The slot illegal instruction exception handling may be generated in this LSI in case where the instruction is described in assembler or when the middleware of the object is introduced. Note that a check-up program (checker) to pick up this instruction is available on our website. Download and utilize this checker as needed.
Rev. 6.00 Jun. 12, 2007 Page 80 of 610 REJ09B0131-0600
Section 6 Interrupt Controller (INTC)
Section 6 Interrupt Controller (INTC)
The interrupt controller (INTC) ascertains the priority of interrupt sources and controls interrupt requests to the CPU.
6.1
Features
* 16 levels of interrupt priority Figure 6.1 shows a block diagram of the INTC.
Rev. 6.00 Jun. 12, 2007 Page 81 of 610 REJ09B0131-0600
Section 6 Interrupt Controller (INTC)
NMI IRQ0
. . .
. . .
Input control
Comparator
IRQ7 UBC H-UDI WDT E-DMAC CMT0 CMT1 SCIF0 SCIF1 SCIF2 HIF (Interrupt request) (Interrupt request) (Interrupt request) (Interrupt request) (Interrupt request) (Interrupt request) (Interrupt request) (Interrupt request) (Interrupt request) (Interrupt request) Priority determination
Interrupt request SR I3 I2 I1 I0 CPU
DTER ICR0 IRQCR IRQSR IPRA to IPRE
Internal bus
IPR
DTC
Module bus
Bus interface
INTC [Legend] UBC: H-UDI: WDT: E-DMAC: CMT: SCIF: User break controller User debugging interface Watchdog timer DMAC for Ethernet controller Compare match timer Serial communications interface with FIFO HIF: ICR0: IRQCR: IRQSR: IPRA to IPRE: SR: Host interface Interrupt control register 0 IRQ control register IRQ status register Interrupt priority registers A to E Status register
Figure 6.1 INTC Block Diagram
Rev. 6.00 Jun. 12, 2007 Page 82 of 610 REJ09B0131-0600
Section 6 Interrupt Controller (INTC)
6.2
Input/Output Pins
Table 6.1 shows the INTC pin configuration. Table 6.1
Name Non-maskable interrupt input pin Interrupt request input pins
Pin Configuration
Abbr. NMI IRQ0 to IRQ7 I/O Input Input Function Input of non-maskable interrupt request signal Input of maskable interrupt request signals
6.3
Register Descriptions
The interrupt controller has the following registers. For details on the addresses of these registers and the states of these registers in each processing state, see section 20, List of Registers. * * * * * * * * Interrupt control register 0 (ICR0) IRQ control register (IRQCR) IRQ status register (IRQSR) Interrupt priority register A (IPRA) Interrupt priority register B (IPRB) Interrupt priority register C (IPRC) Interrupt priority register D (IPRD) Interrupt priority register E (IPRE)
Rev. 6.00 Jun. 12, 2007 Page 83 of 610 REJ09B0131-0600
Section 6 Interrupt Controller (INTC)
6.3.1
Interrupt Control Register 0 (ICR0)
ICR0 is a 16-bit register that sets the input signal detection mode of the external interrupt input pin NMI and indicates the input signal level on the NMI pin.
Bit 15 Initial Bit Name Value NMIL 1/0 R/W R Description NMI Input Level Indicates the state of the signal input to the NMI pin. This bit can be read to determine the NMI pin level. This bit cannot be modified. 0: State of the NMI input is low 1: State of the NMI input is high 14 to 9 All 0 R Reserved These bits are always read as 0. The write value should always be 0. 8 NMIE 0 R/W NMI Edge Select 0: Interrupt request is detected on the falling edge of the NMI input 1: Interrupt request is detected on the rising edge of the NMI input 7 to 0 All 0 R Reserved These bits are always read as 0. The write value should always be 0.
Rev. 6.00 Jun. 12, 2007 Page 84 of 610 REJ09B0131-0600
Section 6 Interrupt Controller (INTC)
6.3.2
IRQ Control Register (IRQCR)
IRQCR is a 16-bit register that sets the input signal detection mode of the external interrupt input pins IRQ0 to IRQ7.
Bit 15 14 Bit Name IRQ71S IRQ70S Initial Value 0 0 R/W R/W R/W Description IRQ7 Sense Select Set the interrupt request detection mode for pin IRQ7. 00: Interrupt request is detected at the low level of pin IRQ7 01: Interrupt request is detected at the falling edge of pin IRQ7 10: Interrupt request is detected at the rising edge of pin IRQ7 11: Interrupt request is detected at both the falling and rising edges of pin IRQ7 13 12 IRQ61S IRQ60S 0 0 R/W R/W IRQ6 Sense Select Set the interrupt request detection mode for pin IRQ6. 00: Interrupt request is detected at the low level of pin IRQ6 01: Interrupt request is detected at the falling edge of pin IRQ6 10: Interrupt request is detected at the rising edge of pin IRQ6 11: Interrupt request is detected at both the falling and rising edges of pin IRQ6
Rev. 6.00 Jun. 12, 2007 Page 85 of 610 REJ09B0131-0600
Section 6 Interrupt Controller (INTC)
Bit 11 10
Bit Name IRQ51S IRQ50S
Initial Value 0 0
R/W R/W R/W
Description IRQ5 Sense Select Set the interrupt request detection mode for pin IRQ5. 00: Interrupt request is detected at the low level of pin IRQ5 01: Interrupt request is detected at the falling edge of pin IRQ5 10: Interrupt request is detected at the rising edge of pin IRQ5 11: Interrupt request is detected at both the falling and rising edges of pin IRQ5
9 8
IRQ41S IRQ40S
0 0
R/W R/W
IRQ4 Sense Select Set the interrupt request detection mode for pin IRQ4. 00: Interrupt request is detected at the low level of pin IRQ4 01: Interrupt request is detected at the falling edge of pin IRQ4 10: Interrupt request is detected at the rising edge of pin IRQ4 11: Interrupt request is detected at both the falling and rising edges of pin IRQ4
7 6
IRQ31S IRQ30S
0 0
R/W R/W
IRQ3 Sense Select Set the interrupt request detection mode for pin IRQ3. 00: Interrupt request is detected at the low level of pin IRQ3 01: Interrupt request is detected at the falling edge of pin IRQ3 10: Interrupt request is detected at the rising edge of pin IRQ3 11: Interrupt request is detected at both the falling and rising edges of pin IRQ3
Rev. 6.00 Jun. 12, 2007 Page 86 of 610 REJ09B0131-0600
Section 6 Interrupt Controller (INTC)
Bit 5 4
Bit Name IRQ21S IRQ20S
Initial Value 0 0
R/W R/W R/W
Description IRQ2 Sense Select Set the interrupt request detection mode for pin IRQ2. 00: Interrupt request is detected at the low level of pin IRQ2 01: Interrupt request is detected at the falling edge of pin IRQ2 10: Interrupt request is detected at the rising edge of pin IRQ2 11: Interrupt request is detected at both the falling and rising edges of pin IRQ2
3 2
IRQ11S IRQ10S
0 0
R/W R/W
IRQ1 Sense Select Set the interrupt request detection mode for pin IRQ1. 00: Interrupt request is detected at the low level of pin IRQ1 01: Interrupt request is detected at the falling edge of pin IRQ1 10: Interrupt request is detected at the rising edge of pin IRQ1 11: Interrupt request is detected at both the falling and rising edges of pin IRQ1
1 0
IRQ01S IRQ00S
0 0
R/W R/W
IRQ0 Sense Select Set the interrupt request detection mode for pin IRQ0. 00: Interrupt request is detected at the low level of pin IRQ0 01: Interrupt request is detected at the falling edge of pin IRQ0 10: Interrupt request is detected at the rising edge of pin IRQ0 11: Interrupt request is detected at both the falling and rising edges of pin IRQ0
Rev. 6.00 Jun. 12, 2007 Page 87 of 610 REJ09B0131-0600
Section 6 Interrupt Controller (INTC)
6.3.3
IRQ Status register (IRQSR)
IRQSR is a 16-bit register that indicates the states of the external interrupt input pins IRQ0 to IRQ7 and the status of interrupt request.
Bit 15 Bit Name IRQ7L Initial Value 0/1 R/W R Description Indicates the state of pin IRQ7. 0: State of pin IRQ7 is low 1: State of pin IRQ7 is high 14 IRQ6L 0/1 R Indicates the state of pin IRQ6. 0: State of pin IRQ6 is low 1: State of pin IRQ6 is high 13 IRQ5L 0/1 R Indicates the state of pin IRQ5. 0: State of pin IRQ5 is low 1: State of pin IRQ5 is high 12 IRQ4L 0 or 1 R Indicates the state of pin IRQ4. 0: State of pin IRQ4 is low 1: State of pin IRQ4 is high 11 IRQ3L 0 or 1 R Indicates the state of pin IRQ3. 0: State of pin IRQ3 is low 1: State of pin IRQ3 is high 10 IRQ2L 0 or 1 R Indicates the state of pin IRQ2. 0: State of pin IRQ2 is low 1: State of pin IRQ2 is high 9 IRQ1L 0 or 1 R Indicates the state of pin IRQ1. 0: State of pin IRQ1 is low 1: State of pin IRQ1 is high 8 IRQ0L 0 or 1 R Indicates the state of pin IRQ0. 0: State of pin IRQ0 is low 1: State of pin IRQ0 is high
Rev. 6.00 Jun. 12, 2007 Page 88 of 610 REJ09B0131-0600
Section 6 Interrupt Controller (INTC)
Bit 7
Bit Name IRQ7F
Initial Value 0
R/W R/W
Description Indicates the status of an IRQ7 interrupt request. * When level detection mode is selected [Clearing condition] Driving pin IRQ7 high 1: An IRQ7 interrupt has been detected [Setting condition] Driving pin IRQ7 low * When edge detection mode is selected [Clearing conditions] Writing 0 after reading IRQ7F = 1 Accepting an IRQ7 interrupt 1: An IRQ7 interrupt request has been detected [Setting condition] Detecting the specified edge of pin IRQ7 0: An IRQ7 interrupt has not been detected 0: An IRQ7 interrupt has not been detected
6
IRQ6F
0
R/W
Indicates the status of an IRQ6 interrupt request. * When level detection mode is selected [Clearing condition] Driving pin IRQ6 high 1: An IRQ6 interrupt has been detected [Setting condition] Driving pin IRQ6 low * When edge detection mode is selected [Clearing conditions] Writing 0 after reading IRQ6F = 1 Accepting an IRQ6 interrupt 1: An IRQ6 interrupt request has been detected [Setting condition] Detecting the specified edge of pin IRQ6 0: An IRQ6 interrupt has not been detected 0: An IRQ6 interrupt has not been detected
Rev. 6.00 Jun. 12, 2007 Page 89 of 610 REJ09B0131-0600
Section 6 Interrupt Controller (INTC)
Bit 5
Bit Name IRQ5F
Initial Value 0
R/W R/W
Description Indicates the status of an IRQ5 interrupt request. * When level detection mode is selected [Clearing condition] Driving pin IRQ5 high 1: An IRQ5 interrupt has been detected [Setting condition] Driving pin IRQ5 low * When edge detection mode is selected [Clearing conditions] Writing 0 after reading IRQ5F = 1 Accepting an IRQ5 interrupt 1: An IRQ5 interrupt request has been detected [Setting condition] Detecting the specified edge of pin IRQ5 0: An IRQ5 interrupt has not been detected 0: An IRQ5 interrupt has not been detected
4
IRQ4F
0
R/W
Indicates the status of an IRQ4 interrupt request. * When level detection mode is selected [Clearing condition] Driving pin IRQ4 high 1: An IRQ4 interrupt has been detected [Setting condition] Driving pin IRQ4 low * When edge detection mode is selected [Clearing conditions] Writing 0 after reading IRQ4F = 1 Accepting an IRQ4 interrupt 1: An IRQ4 interrupt request has been detected [Setting condition] Detecting the specified edge of pin IRQ4 0: An IRQ4 interrupt has not been detected 0: An IRQ4 interrupt has not been detected
Rev. 6.00 Jun. 12, 2007 Page 90 of 610 REJ09B0131-0600
Section 6 Interrupt Controller (INTC)
Bit 3
Bit Name IRQ3F
Initial Value 0
R/W R/W
Description Indicates the status of an IRQ3 interrupt request. * When level detection mode is selected [Clearing condition] Driving pin IRQ3 high 1: An IRQ3 interrupt has been detected [Setting condition] Driving pin IRQ3 low * When edge detection mode is selected [Clearing conditions] Writing 0 after reading IRQ3F = 1 Accepting an IRQ3 interrupt 1: An IRQ3 interrupt request has been detected [Setting condition] Detecting the specified edge of pin IRQ3 0: An IRQ3 interrupt has not been detected 0: An IRQ3 interrupt has not been detected
2
IRQ2F
0
R/W
Indicates the status of an IRQ2 interrupt request. * When level detection mode is selected [Clearing condition] Driving pin IRQ2 high 1: An IRQ2 interrupt has been detected [Setting condition] Driving pin IRQ2 low * When edge detection mode is selected [Clearing conditions] Writing 0 after reading IRQ2F = 1 Accepting an IRQ2 interrupt 1: An IRQ2 interrupt request has been detected [Setting condition] Detecting the specified edge of pin IRQ2 0: An IRQ2 interrupt has not been detected 0: An IRQ2 interrupt has not been detected
Rev. 6.00 Jun. 12, 2007 Page 91 of 610 REJ09B0131-0600
Section 6 Interrupt Controller (INTC)
Bit 1
Bit Name IRQ1F
Initial Value 0
R/W R/W
Description Indicates the status of an IRQ1 interrupt request. * When level detection mode is selected [Clearing condition] Driving pin IRQ1 high 1: An IRQ1 interrupt has been detected [Setting condition] Driving pin IRQ1 low * When edge detection mode is selected [Clearing conditions] Writing 0 after reading IRQ1F = 1 Accepting an IRQ1 interrupt 1: An IRQ1 interrupt request has been detected [Setting condition] Detecting the specified edge of pin IRQ1 0: An IRQ1 interrupt has not been detected 0: An IRQ1 interrupt has not been detected
0
IRQ0F
0
R/W
Indicates the status of an IRQ0 interrupt request. * When level detection mode is selected [Clearing condition] Driving pin IRQ0 high 1: An IRQ0 interrupt has been detected [Setting condition] Driving pin IRQ0 low * When edge detection mode is selected [Clearing conditions] Writing 0 after reading IRQ0F = 1 Accepting an IRQ0 interrupt 1: An IRQ0 interrupt request has been detected [Setting condition] Detecting the specified edge of pin IRQ0 0: An IRQ0 interrupt has not been detected 0: An IRQ0 interrupt has not been detected
Rev. 6.00 Jun. 12, 2007 Page 92 of 610 REJ09B0131-0600
Section 6 Interrupt Controller (INTC)
6.3.4
Interrupt Priority Registers A to E (IPRA to IPRE)
Interrupt priority registers are five 16-bit readable/writable registers that set priority levels from 0 to 15 for interrupts except NMI. For the correspondence between interrupt request sources and IPR, refer to table 6.2, Interrupt Exception Handling Vectors and Priorities. Each of the corresponding interrupt priority ranks are established by setting a value from H'0 to H'F in each of the four-bit groups 15 to 12, 11 to 8, 7 to 4 and 3 to 0. Reserved bits that are not assigned should be set H'0 (B'0000).
Bit 15 14 13 12 Bit Name IPR15 IPR14 IPR13 IPR12 Initial Value 0 0 0 0 R/W R/W R/W R/W R/W Description Set priority levels for the corresponding interrupt source. 0000: 0001: 0010: 0011: 0100: 0101: 0110: 0111: 1000: 1001: 1010: 1011: 1100: 1101: 1110: 1111: Priority level 0 (lowest) Priority level 1 Priority level 2 Priority level 3 Priority level 4 Priority level 5 Priority level 6 Priority level 7 Priority level 8 Priority level 9 Priority level 10 Priority level 11 Priority level 12 Priority level 13 Priority level 14 Priority level 15 (highest)
Rev. 6.00 Jun. 12, 2007 Page 93 of 610 REJ09B0131-0600
Section 6 Interrupt Controller (INTC)
Bit 11 10 9 8
Bit Name IPR11 IPR10 IPR9 IPR8
Initial Value 0 0 0 0
R/W R/W R/W R/W R/W
Description Set priority levels for the corresponding interrupt source. 0000: 0001: 0010: 0011: 0100: 0101: 0110: 0111: 1000: 1001: 1010: 1011: 1100: 1101: 1110: 1111: Priority level 0 (lowest) Priority level 1 Priority level 2 Priority level 3 Priority level 4 Priority level 5 Priority level 6 Priority level 7 Priority level 8 Priority level 9 Priority level 10 Priority level 11 Priority level 12 Priority level 13 Priority level 14 Priority level 15 (highest)
7 6 5 4
IPR7 IPR6 IPR5 IPR4
0 0 0 0
R/W R/W R/W R/W
Set priority levels for the corresponding interrupt source. 0000: 0001: 0010: 0011: 0100: 0101: 0110: 0111: 1000: 1001: 1010: 1011: 1100: 1101: 1110: 1111: Priority level 0 (lowest) Priority level 1 Priority level 2 Priority level 3 Priority level 4 Priority level 5 Priority level 6 Priority level 7 Priority level 8 Priority level 9 Priority level 10 Priority level 11 Priority level 12 Priority level 13 Priority level 14 Priority level 15 (highest)
Rev. 6.00 Jun. 12, 2007 Page 94 of 610 REJ09B0131-0600
Section 6 Interrupt Controller (INTC)
Bit 3 2 1 0
Bit Name IPR3 IPR2 IPR1 IPR0
Initial Value 0 0 0 0
R/W R/W R/W R/W R/W
Description Set priority levels for the corresponding interrupt source. 0000: 0001: 0010: 0011: 0100: 0101: 0110: 0111: 1000: 1001: 1010: 1011: 1100: 1101: 1110: 1111: Priority level 0 (lowest) Priority level 1 Priority level 2 Priority level 3 Priority level 4 Priority level 5 Priority level 6 Priority level 7 Priority level 8 Priority level 9 Priority level 10 Priority level 11 Priority level 12 Priority level 13 Priority level 14 Priority level 15 (highest)
Note: Name in the tables above is represented by a general name. Name in the list of register is, on the other hand, represented by a module name.
6.4
6.4.1
Interrupt Sources
External Interrupts
There are five types of interrupt sources: User break, NMI, H-UDI, IRQ, and on-chip peripheral modules. Individual interrupts are given priority levels (0 to 16, with 0 the lowest and 15 the highest). Giving an interrupt a priority level of 0 masks it. NMI Interrupt: The NMI interrupt is given a priority level of 16 and is always accepted. An NMI interrupt is detected at the edge of the pins. Use the NMI edge select bit (NMIE) in interrupt control register 0 (ICR0) to select either the rising or falling edge. In the NMI interrupt exception handler, the interrupt mask level bits (I3 to I0) in the status register (SR) are set to level 15.
Rev. 6.00 Jun. 12, 2007 Page 95 of 610 REJ09B0131-0600
Section 6 Interrupt Controller (INTC)
IRQ7 to IRQ0 Interrupts: IRQ interrupts are requested by input from pins IRQ0 to IRQ7. Use the IRQ sense select bits (IRQ71S to IRQ 01S and IRQ70S to IRQ00S) in the IRQ control register (IRQCR) to select the detection mode from low level detection, falling edge detection, rising edge detection, and both edge detection for each pin. The priority level can be set from 0 to 15 for each pin using the interrupt priority registers A and B (IPRA and IPRB). In the case that the low level detection is selected, an interrupt request signal is sent to the INTC while the IRQ pin is driven low. The interrupt request signal stops to be sent to the INTC when the IRQ pin becomes high. It is possible to confirm that an interrupt is requested by reading the IRQ flags (IRQ7F to IRQ0F) in the IRQ status register (IRQSR). In the case that the edge detection is selected, an interrupt request signal is sent to the INTC when the following change on the IRQ pin is detected: from high to low in falling edge detection mode, from low to high in rising edge detection mode, and from low to high or from high to low in both edge detection mode. The IRQ interrupt request by detecting the change on the pin is held until the interrupt request is accepted. It is possible to confirm that an IRQ interrupt request has been detected by reading the IRQ flags (IRQ7F to IRQ0F) in the IRQ status register (IRQSR). An IRQ interrupt request by detecting the change on the pin can be withdrawn by writing 0 to an IRQ flag after reading 1. In the IRQ interrupt exception handling, the interrupt mask bits (I3 to I0) in the status register (SR) are set to the priority level value of the accepted IRQ interrupt. Figure 6.2 shows the block diagram of the IRQ7 to IRQ0 interrupts.
IRQSR.IRQnL IRQCR.IRQn1S IRQCR.IRQn0S IRQSR.IRQnF
Selection
IRQn pins
Level detection Edge detection S Q
CPU interrupt request
RESIRQn (Acceptance of IRQn interrupt/ writing 0 after reading IRQnF = 1)
R
n = 7 to 0
Figure 6.2 Block Diagram of IRQ7 to IRQ0 Interrupts Control
Rev. 6.00 Jun. 12, 2007 Page 96 of 610 REJ09B0131-0600
Section 6 Interrupt Controller (INTC)
6.4.2
On-Chip Peripheral Module Interrupts
On-chip peripheral module interrupts are interrupts generated by the following on-chip peripheral modules. Since a different interrupt vector is allocated to each interrupt source, the exception handling routine does not have to decide which interrupt has occurred. Priority levels between 0 and 15 can be allocated to individual on-chip peripheral modules in interrupt priority registers C to E (IPRC to IPRE). On-chip peripheral module interrupt exception handling sets the interrupt mask level bits (I3 to I0) in the status register (SR) to the priority level value of the on-chip peripheral module interrupt that was accepted. 6.4.3 User Break Interrupt
A user break interrupt has a priority level of 15, and occurs when the break condition set in the user break controller (UBC) is satisfied. User break interrupt requests are detected by edge and are held until accepted. User break interrupt exception handling sets the interrupt mask level bits (I3 to I0) in the status register (SR) to level 15. For more details on the user break interrupt, see section 18, User Break Controller (UBC). 6.4.4 H-UDI Interrupt
User debugging interface (H-UDI) interrupt has a priority level of 15, and occurs when an H-UDI interrupt instruction is serially input. H-UDI interrupt requests are detected by edge and are held until accepted. H-UDI exception handling sets the interrupt mask level bits (I3-I0) in the status register (SR) to level 15. For more details on the H-UDI interrupt, see section 19, User Debugging Interface (H-UDI).
Rev. 6.00 Jun. 12, 2007 Page 97 of 610 REJ09B0131-0600
Section 6 Interrupt Controller (INTC)
6.5
Interrupt Exception Handling Vector Table
Table 6.2 lists interrupt sources, their vector numbers, vector table address offsets, and interrupt priorities. Individual interrupt sources are allocated to different vector numbers and vector table address offsets. Vector table addresses are calculated from the vector numbers and vector table address offsets. For interrupt exception handling, the start address of the exception handling routine is fetched from the vector table address in the vector table. For the details on calculation of vector table addresses, see table 5.4, Calculating Exception Handling Vector Table Addresses in section 5, Exception Handling. IRQ interrupts and on-chip peripheral module interrupt priorities can be set freely between 0 and 15 for each pin or module by setting interrupt priority registers A to E (IPRA to IPRE). However, when interrupt sources whose priority levels are allocated with the same IPR are requested, the interrupt of the smaller vector number has priority. This priority cannot be changed. Priority levels of IRQ interrupts and on-chip peripheral module interrupts are initialized to level 0 at a power-on reset. If the same priority level is allocated to two or more interrupt sources and interrupts from those sources occur simultaneously, they are processed by the default priority order shown in table 6.2. Table 6.2
Interrupt Source User break External pin H-UDI External pin IRQ0 IRQ1 IRQ2 IRQ3 IRQ4 IRQ5 IRQ6 IRQ7 WDT ITI NMI
Interrupt Exception Handling Vectors and Priorities
Name Vector No. 12 11 13 64 65 66 67 80 81 82 83 84 Vector Table Starting Address H'00000030 H'0000002C H'00000034 H'00000100 H'00000104 H'00000108 H'0000010C H'00000140 H'00000144 H'00000148 H'0000014C H'00000150 IPR IPRA15 to IPRA12 IPRA11 to IPRA8 IPRA7 to IPRA4 IPRA3 to IPRA0 IPRB15 to IPRB12 IPRB11 to IPRB8 IPRB7 to IPRB4 IPRB3 to IPRB0 IPRC15 to IPRC12 Low Default Priority High
Rev. 6.00 Jun. 12, 2007 Page 98 of 610 REJ09B0131-0600
Section 6 Interrupt Controller (INTC)
Interrupt Source E-DMAC
Name EINT0
Vector No. 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101
Vector Table Starting Address H'00000154 H'00000158 H'0000015C H'00000160 H'00000164 H'00000168 H'0000016C H'00000170 H'00000174 H'00000178 H'0000017C H'00000180 H'00000184 H'00000188 H'0000018C H'00000190 H'00000194
IPR IPRC11 to IPRC8 IPRC7 to IPRC4 IPRC3 to IPRC0 IPRD15 to IPRD12
Default Priority High
CMT channel 0 CMI0 CMT channel 1 CMI1 SCIF channel 0 ERI_0 RXI_0 BRI_0 TXI_0 SCIF channel 1 ERI_1 RXI_1 BRI_1 TXI_1 SCIF channel 2 ERI_2 RXI_2 BRI_2 TXI_2 HIF HIFI HIFBI
IPRD11 to IPRD8
IPRD7 to IPRD4
IPRE15 to IPRE12 IPRE11 to IPRE8 Low
Rev. 6.00 Jun. 12, 2007 Page 99 of 610 REJ09B0131-0600
Section 6 Interrupt Controller (INTC)
6.6
6.6.1
Interrupt Operation
Interrupt Sequence
The sequence of interrupt operations is explained below. Figure 6.3 is a flowchart of the operations. 1. The interrupt request sources send interrupt request signals to the interrupt controller. 2. The interrupt controller selects the highest priority interrupt from interrupt requests sent, according to the priority levels set in interrupt priority level setting registers A to E (IPRA to IPRE). Interrupts that have lower-priority than that of the selected interrupt are ignored*. If interrupts that have the same priority level or interrupts within a same module occur simultaneously, the interrupt with the highest priority is selected according to the priority shown in table 6.2. 3. The interrupt controller compares the priority level of the selected interrupt request with the interrupt mask bits (I3 to I0) in the status register (SR) of the CPU. If the priority level of the selected request is equal to or less than the level set in bits I3 to I0, the request is ignored. If the priority level of the selected request is higher than the level in bits I3 to I0, the interrupt controller accepts the request and sends an interrupt request signal to the CPU. 4. The CPU detects the interrupt request sent from the interrupt controller in the decode stage of an instruction to be executed. Instead of executing the decoded instruction, the CPU starts interrupt exception handling (see figure 6.5). 5. SR and PC are saved onto the stack. 6. The priority level of the accepted interrupt is copied to bits (I3 to I0) in SR. 7. The CPU reads the start address of the exception handling routine from the exception vector table for the accepted interrupt, branches to that address, and starts executing the program. This branch is not a delayed branch. Note: * Interrupt requests that are designated as edge-detect type are held pending until the interrupt requests are accepted. IRQ interrupts, however, can be cancelled by accessing the IRQ status register (IRQSR). Interrupts held pending due to edge detection are cleared by a power-on reset or an H-UDI reset.
Rev. 6.00 Jun. 12, 2007 Page 100 of 610 REJ09B0131-0600
Section 6 Interrupt Controller (INTC)
Program execution state
Interrupt? Yes User break? Yes
No
No
NMI? Yes
No
H-UDI interrupt? Yes
No
Level 15 interrupt? Yes
No
I3 to I0 level 14? Yes
Yes No
I3 to I0 level 14? No Yes
Level 14 interrupt? Yes I3 to I0 level 13? No Yes
No
Level 1 interrupt? Yes I3 to I0 = level 0? No
No
Save SR to stack Save PC to stack Copy interrupt level to I3 to I0 Read exception vector table Branch to exception handling routine
Note: I3 to I0 are Interrupt mask bits in the status register (SR) of the CPU
Figure 6.3 Interrupt Sequence Flowchart
Rev. 6.00 Jun. 12, 2007 Page 101 of 610 REJ09B0131-0600
Section 6 Interrupt Controller (INTC)
6.6.2
Stack after Interrupt Exception Handling
Figure 6.4 shows the stack after interrupt exception handling.
Address 4n - 8 4n - 4 4n PC*1 SR 32 bits 32 bits SP*2
Notes: 1. PC is the start address of the next instruction (instruction at the return address) after the executed instruction. 2. Always make sure that SP is a multiple of 4
Figure 6.4 Stack after Interrupt Exception Handling
6.7
Interrupt Response Time
Table 6.3 lists the interrupt response time, which is the time from the occurrence of an interrupt request until the interrupt exception handling starts and fetching of the first instruction of the interrupt handling routine begins. Figure 6.5 shows an example of the pipeline operation when an IRQ interrupt is accepted.
Rev. 6.00 Jun. 12, 2007 Page 102 of 610 REJ09B0131-0600
Section 6 Interrupt Controller (INTC)
Table 6.3
Interrupt Response Time
Number of Cycles
Item Interrupt priority decision and comparison with mask bits in SR Wait for completion of sequence currently being executed by CPU
NMI, H-UDI 1 x Icyc + 2 x Pcyc
IRQ, Peripheral Modules 1 x Icyc + 3 x Pcyc
Remarks
X ( 0)
X ( 0)
The longest sequence is for interrupt or address-error exception handling (X = 7 x Icyc + m1 + m2 + m3 + m4). If an interrupt-masking instruction follows, however, the time may be even longer. Performs the saving PC and SR, and vector address fetch.
Time from start of interrupt exception handling until fetch of first instruction of exception handling routine starts Interrupt response time Total:
8 x Icyc + m1 + m2 + m3
8 x Icyc + m1 + m2 + m3
9 x Icyc + 2 x Pcyc + m1 + m2 + m3 +X 12 x Icyc + 2 x Pcyc
9 x Icyc + 3 x Pcyc + m1 + m2 + m3 +X 12 x Icyc + 3 x Pcyc SR, PC, and vector table are all in on-chip RAM, or cache hit occurs (in write back mode).
Minimum*:
Maximum:
16 x Icyc + 2 x Pcyc + 2 x (m1 + m2 + m3) + m4
16 x Icyc + 3 x Pcyc + 2 x (m1 + m2 + m3) + m4
Notes: *
In the case that m1 = m2 = m3 = m4 = 1 x Icyc. m1 to m4 are the number of cycles needed for the following memory accesses. m1: SR save (longword write) m2: PC save (longword write) m3: Vector address read (longword read) m4: Fetch first instruction of interrupt service routine
Rev. 6.00 Jun. 12, 2007 Page 103 of 610 REJ09B0131-0600
Section 6 Interrupt Controller (INTC)
Rev. 6.00 Jun. 12, 2007 Page 104 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Section 7 Bus State Controller (BSC)
The bus state controller (BSC) outputs control signals for various types of memory that is connected to the external address space and external devices. The BSC functions enable this LSI to connect directly with SRAM, SDRAM, and other memory storage devices, and external devices.
7.1
Features
The BSC has the following features. * External address space A maximum 32 or 64 Mbytes for each of the areas, CS0, CS3, CS4, CS5B, and CS6B, totally 256 Mbytes (divided into five areas) A maximum 64 Mbytes for each of the six areas, CS0, CS3, CS4, CS5, and CS6, totally 320 Mbytes (divided into five areas) Can specify the normal space interface, byte-selection SRAM, SDRAM, PCMCIA for each address space Can select the data bus width (8 or 16 bits) for each address space Can control the insertion of wait cycles for each address space Can control the insertion of wait cycles for each read access and write access Can control the insertion of idle cycles in the consecutive access for five cases independently: read-write (in same space/different space), read-read (in same space/different space), or the first cycle is a write access * Normal space interface Supports the interface that can directly connect to the SRAM * SDRAM interface Can connect directly to SDRAM in area 3 Multiplex output for row address/column address Efficient access by single read/single write High-speed access by bank-active mode Supports auto-refreshing and self-refreshing * Byte-selection SRAM interface Can connect directly to byte-selection SRAM
Rev. 6.00 Jun. 12, 2007 Page 105 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
* PCMCIA direct interface Supports IC memory cards and I/O card interfaces defined in the JEIDA specifications Ver 4.2 (PCMCIA2.1 Rev 2.1) Controls the insertion of wait cycles by software Supports the bus sizing function of the I/O bus width (only in little endian mode) * Refresh function Supports the auto-refreshing and self-refreshing functions Specifies the refresh interval by setting the refresh counter and clock selection Can execute consecutive refresh cycles by specifying the refresh counts (1, 2, 4, 6, or 8)
Rev. 6.00 Jun. 12, 2007 Page 106 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
The block diagram of the BSC is shown in figure 7.1.
CMNCR
Internal bus
Bus mastership controller
Internal master module
Internal slave module
CS0WCR WAIT Wait controller
...
...
CS6BWCR RWTCNT CS0, CS3, CS4, CS5B (CE1A), CS6B (CE1B) MD5 A25 to A0, D15 to D0, BS, RD/WR, RD, WE1 (BE1, DQMLU, WE), WE0 (BE0, DQMLL), ICIOWR, ICIORD, RAS, CAS, CKE, CE2A, CE2B
CS6BBCR
Memory controller
IOIS16
SDCR RTCSR RTCNT Refresh controller
Comparator
RTCOR [Legend] CMNCR: CSnWCR: RWTCNT: CSnBCR: SDCR: RTCSR: RTCNT: RTCOR: BSC Common control register CSn space wait control register (n = 0, 3, 4, 5B, 6B) Reset wait counter CSn space bus control register (n = 0, 3, 4, 5B, 6B) SDRAM control register Refresh timer control/status register Refresh timer counter Refresh time constant register
Figure 7.1 Block Diagram of BSC
Rev. 6.00 Jun. 12, 2007 Page 107 of 610 REJ09B0131-0600
Module bus
Area controller
CS0BCR
... ...
...
Section 7 Bus State Controller (BSC)
7.2
Input/Output Pins
Table 7.1 lists the pin configuration of the BSC. Table 7.1 Pin Configuration
I/O Function
Abbreviation A25 to A0 D15 to D0 BS
Output Address Bus* I/O Data Bus Asserted when a normal space, burst ROM (clock synchronous /asynchronous), or PCMCIA is accessed. Asserted at the same timing as CAS assertion in SDRAM access.
Output Bus Cycle Start
CS0, CS3, CS4 Output Chip Select CS5B/CE1A CE2A CS6B/CE1B CE2B RD/WR RD WE1(BE1)/WE Output Chip Select Chip enable for PCMCIA allocated to area 5 when PCMCIA is in use Output Chip enable for PCMCIA allocated to area 5 when PCMCIA is in use Output Chip Select Chip enable for PCMCIA allocated to area 6 when PCMCIA is in use Output Chip enable for PCMCIA allocated to area 6 when PCMCIA is in use Output Read/Write Connects to WE pins when SDRAM or byte-selection SRAM is used. Output Read Pulse Signal (read data output enable signal) Strobe signal to indicate a memory read cycle when PCMCIA is in use. Output Indicates that D15 to D8 are being written to. Connected to the byte select signal when byte-selection SRAM is in use. Strove signal to indicate a memory write cycle when PCMCIA is in use. WE0(BE0) Output Indicates that D7 to D0 are being written to. Connected to the byte select signal when a byte-selection SRAM is in use. RAS CAS CKE Output Connected to RAS pin when SDRAM is in use. Output Connected to CAS pin when SDRAM is in use. Output Connected to CKE pin when SDRAM is in use.
Rev. 6.00 Jun. 12, 2007 Page 108 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Abbreviation IOIS16
I/O Input
Function PCMCIA 16-bit I/O Signal Enabled only in little endian mode. Drive this signal low in big endian mode.
DQMLU, DQMLL
Output Connected to the DQMxx pin when SDRAM is in use. DQMLU: Select signal for D15 to D8 DQMLL: Select signal for D7 to D0 Input Input External wait input MD5: Selects data alignment (big endian or little endian) MD3: Specifies area 0 bus width (8/16 bits)
WAIT MD5, MD3 Note: *
As pins A25 to A16 act as general I/O ports immediately after a power-on reset, pull up or pull down these pins outside the LSI as needed.
7.3
7.3.1
Area Overview
Area Division
The architecture of this LSI has 32-bit address space. The upper three address bits divide the space into areas P0 to P4, and the cache access methods can be specified for each area. For details, see section 3, Cache. Each area indicated by the remaining 29 bits is divided into ten areas (five areas are reserved) when address map 1 is selected or eight areas (three areas are reserved) when address map 2 is selected. The address map is selected by the MAP bit in CMNCR. The BSC controls the areas indicated by the 29 bits. As listed in tables 7.2 and 7.3, memory can be connected directly to five physical areas of this LSI, and the chip select signals (CS0, CS3, CS4, CS5B, and CS6B) are output for each area. CS0 is asserted during area 0 access. 7.3.2 Shadow Area
Areas 0, 3, 4, 5B, and 6B are divided by decoding physical address bits A28 to A25, which correspond to areas 000 to 111. Address bits 31 to 29 are ignored. This means that the range of area 0 addresses, for example, is H'00000000 to H'03FFFFFF, and its corresponding shadow space is the address space in P1 to P3 areas obtained by adding to it H'20000000 x n (n = 1 to 6). The address range for area 7 is H'1C000000 to H'1FFFFFFF. The address space H'1C000000 + H'20000000 x n to H'1FFFFFFF + H'20000000 x n (n = 0 to 6) corresponding to the area 7 shadow spaces are reserved, so do not use it.
Rev. 6.00 Jun. 12, 2007 Page 109 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Area P4 (H'E0000000 to H'EFFFFFFF) is an I/O area and is allocated to internal register addresses. Therefore, area P4 does not become shadow space.
H'00000000 H'20000000 H'40000000 H'60000000 H'80000000 P1 H'A0000000 P2 H'C0000000 P3 H'E0000000 P4 Address Space P0
Area 0 (CS0) Area 1 (reserved) Area 2 (reserved) Area 3 (CS3) Area 4 (CS4) Area 5A (reserved) Area 5B (CS5B) Area 6A (reserved) Area 6B (CS6B) Area 7 (reserved) Physical address space
Figure 7.2 Address Space 7.3.3 Address Map
The external address space has a capacity of 256 Mbytes and is divided into five areas. Types of memory to be connected and the data bus width are specified for individual areas. The address map for the external address space is shown in table 7.2. Table 7.2 Address Map 1 (CMNCR.MAP = 0)
Area Area 0 Area 1 Area 2 Area 3 Memory to be Connected Normal memory Reserved area* Reserved area* Normal memory Byte-selection SRAM SDRAM Capacity 64 Mbytes 64 Mbytes 64 Mbytes 64 Mbytes
Physical Address H'00000000 to H'03FFFFFF H'04000000 to H'07FFFFFF H'08000000 to H'0BFFFFFF H'0C000000 to H'0FFFFFFF
Rev. 6.00 Jun. 12, 2007 Page 110 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Physical Address H'10000000 to H'13FFFFFF H'14000000 to H'15FFFFFF H'16000000 to H'17FFFFFF H'18000000 to H'19FFFFFF H'1A000000 to H'1BFFFFFF H'1C000000 to H'1FFFFFFF Note: *
Area Area 4 Area 5A Area 5B Area 6A Area 6B Area 7
Memory to be Connected Normal memory Byte-selection SRAM Reserved area* Normal memory Byte-selection SRAM Reserved area* Normal memory Byte-selection SRAM Reserved area*
Capacity 64 Mbytes 32 Mbytes 32 Mbytes 32 Mbytes 32 Mbytes 64 Mbytes
Do not access the reserved area. If the reserved area is accessed, the correct operation cannot be guaranteed.
Table 7.3
Address Map 2 (CMNCR.MAP = 1)
Area Area 0 Area 1 Area 2 Area 3 Memory to be Connected Normal memory Reserved area* Reserved area*
1 1
Physical Address H'00000000 to H'03FFFFFF H'04000000 to H'07FFFFFF H'08000000 to H'0BFFFFFF H'0C000000 to H'0FFFFFFF
Capacity 64 Mbytes 64 Mbytes 64 Mbytes 64 Mbytes
Normal memory Byte-selection SRAM SDRAM
H'10000000 to H'13FFFFFF H'14000000 to H'17FFFFFF
Area 4 Area 5*
2
Normal memory Byte-selection SRAM Normal memory Byte-selection SRAM PCMCIA
64 Mbytes 64 Mbytes
H'18000000 to H'1BFFFFFF
Area 6*2
Normal memory Byte-selection SRAM PCMCIA
64 Mbytes
H'1C000000 to H'1FFFFFFF
Area 7
Reserved area*
1
64 Mbytes
Notes: 1. Do not access the reserved area. If the reserved area is accessed, the correct operation cannot be guaranteed. 2. For area 5, CS5BBCR and CS5BWCR are enabled. For area 6, CS6BBCR and CS6BWCR are enabled.
Rev. 6.00 Jun. 12, 2007 Page 111 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
7.3.4
Area 0 Memory Type and Memory Bus Width
The memory bus width in this LSI can be set for each area. In area 0, the bus width is selected from 8 bits and 16 bits at a power-on reset by the external pin setting. The bus width of other areas is set by the register. The correspondence between the memory type, external pin (MD3), and bus width is listed in table 7.4. Table 7.4
MD3 1 0
Correspondence between External Pin (MD3), Memory Type, and Bus Width for CS0
Memory Type Normal memory Bus Width 8 bits 16 bits
7.3.5
Data Alignment
This LSI supports the big endian and little endian methods of data alignment. The data alignment is specified using the external pin (MD5) at a power-on reset as shown in table 7.5. Table 7.5
MD5 0 1
Correspondence between External Pin (MD5) and Endians
Endian Big endian Little endian
Rev. 6.00 Jun. 12, 2007 Page 112 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
7.4
Register Descriptions
The BSC has the following registers. For the addresses and access size for these registers, see section 20, List of Registers. Do not access spaces other than CS0 until setting the memory interfaces is complete. * * * * * * * * * * * * * * * Common control register (CMNCR) CS0 space bus control register for area 0 (CS0BCR) CS3 space bus control register for area 3 (CS3BCR) CS4 space bus control register for area 4 (CS4BCR) CS5B space bus control register for area 5B (CS5BBCR) CS6B space bus control register for area 6B (CS6BBCR) CS0 space wait control register for area 0 (CS0WCR) CS3 space wait control register for area 3 (CS3WCR) CS4 space wait control register for area 4 (CS4WCR) CS5B space wait control register for area 5B (CS5BWCR) CS6B space wait control register for area 6B (CS6BWCR) SDRAM control register (SDCR) Refresh timer control/status register (RTCSR) Refresh timer counter (RTCNT) Refresh time constant register (RTCOR)
Rev. 6.00 Jun. 12, 2007 Page 113 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
7.4.1
Common Control Register (CMNCR)
CMNCR is a 32-bit register that controls the common items for each area. Do not access external memory other than area 0 until setting CMNCR is complete.
Bit 31 to 13 Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 12 MAP 0 R/W Space Specification Selects the address map for the external address space. The address maps to be selected are shown in tables 7.2 and 7.3. 0: Selects address map 1 1: Selects address map 2 11 to 5 All 0 R Reserved These bits are always read as 0. The write value should always be 0. 4 1 R Reserved This bit is always read as 1. The write value should always be 1. 3 ENDIAN 0/1* R Endian Flag Fetches the external pin (MD5) state for specifying endian at a power-on reset. The endian setting for all the address spaces are set by this bit. This is a read-only bit. 0: External pin (MD5) for specifying endian was driven low at a power-on reset. This LSI is operated as big endian. 1: External pin (MD5) for specifying endian was driven high at a power-on reset. This LSI is being operated as little endian. 2 1 R Reserved This bit is always read as 1. The write value should always be 1.
Rev. 6.00 Jun. 12, 2007 Page 114 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Bit 1
Bit Name HIZMEM
Initial Value 0
R/W R/W
Description Hi-Z Memory Control Specifies the pin state in standby mode for pins A25 to A0, BS, CSn, RD/WR, WEn (BEn)/DQMxx, and RD. 0: High impedance in standby mode 1: Driven in standby mode
0
HIZCNT
0
R/W
Hi-Z Control Specifies the pin state in standby mode for the CKIO, CKE, RAS, and CAS pins. 0: High impedance in standby mode 1: Driven in standby mode
Note:
*
The external pin (MD5) state for specifying endian is sampled at a power-on reset. When big endian is specified, this bit is read as 0 and when little endian is specified, this bit is read as 1.
7.4.2
CSn Space Bus Control Register (CSnBCR) (n = 0, 2, 3, 4, 5B, 6B)
CSnBCR specifies the type of memory connected to each space, data-bus width of each space, and the number of wait cycles between access cycles. Do not access external memory other than area 0 until setting CSnBCR is completed.
Bit 31, 30 Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 29 28 IWW1 IWW0 1 1 R/W R/W Idle Cycles between Write-Read Cycles and Write-Write Cycles Specify the number of idle cycles to be inserted after the access to a memory that is connected to the area. The write and read cycles or write and write cycles performed consecutively are the target cycle. 000: No idle cycle inserted 001: 1 idle cycle inserted 010: 2 idle cycles inserted 011: 4 idle cycles inserted
Rev. 6.00 Jun. 12, 2007 Page 115 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Bit 27
Bit Name
Initial Value 0
R/W R
Description Reserved This bit is always read as 0. The write value should always be 0.
26 25
IWRWD1 IWRWD0
1 1
R/W R/W
Idle Cycles for Another Space Read-Write Specify the number of idle cycles to be inserted after the access to a memory that is connected to the area. The read and write cycles which are performed consecutively and are accessed to different areas are the target cycle. 000: No idle cycle inserted 001: 1 idle cycles inserted 010: 2 idle cycles inserted 011: 4 idle cycles inserted
24
0
R
Reserved This bit is always read as 0. The write value should always be 0.
23 22
IWRWS1 IWRWS0
1 1
R/W R/W
Idle Cycles for Read-Write in Same Space Specify the number of idle cycles to be inserted after the access to a memory that is connected to the area. The read and write cycles which are performed consecutively and are accessed to the same area are the target cycle. 000: No idle cycle inserted 001: 1 idle cycles inserted 010: 2 idle cycles inserted 011: 4 idle cycles inserted
21
0
R
Reserved This bit is always read as 0. The write value should always be 0.
Rev. 6.00 Jun. 12, 2007 Page 116 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Bit 20 19
Bit Name IWRRD1 IWRRD0
Initial Value 1 1
R/W R/W R/W
Description Idle Cycles for Read-Read in Another Space Specify the number of idle cycles to be inserted after the access to a memory that is connected to the area. The read and read cycles which are performed consecutively and are accessed to different areas are the target cycle. 000: No idle cycle inserted 001: 1 idle cycles inserted 010: 2 idle cycles inserted 011: 4 idle cycles inserted
18
0
R
Reserved This bit is always read as 0. The write value should always be 0.
17 16
IWRRS1 IWRRS0
1 1
R/W R/W
Idle Cycles for Read-Read in Same Space Specify the number of idle cycles to be inserted after the access to a memory that is connected to the area. The read and read cycles which are performed consecutively and are accessed to the same area are the target cycle. 000: No idle cycle inserted 001: 1 idle cycles inserted 010: 2 idle cycles inserted 011: 4 idle cycles inserted
Rev. 6.00 Jun. 12, 2007 Page 117 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Bit 15 14 13 12
Bit Name TYPE3 TYPE2 TYPE1 TYPE0
Initial Value 0 0 0 0
R/W R/W R/W R/W R/W
Description Memory Type Specify the type of memory connected to the area. 0000: Normal space 0001: Reserved (setting prohibited) 0010: Reserved (setting prohibited) 0011: Byte-selection SRAM 0100: SDRAM 0101: PCMCIA 0110: Reserved (setting prohibited) 0111: Reserved (setting prohibited) 1000: Reserved (setting prohibited) 1001: Reserved (setting prohibited) 1010: Reserved (setting prohibited) 1011: Reserved (setting prohibited) 1100: Reserved (setting prohibited) 1101: Reserved (setting prohibited) 1110: Reserved (setting prohibited) 1111: Reserved (setting prohibited) For details on memory type in each area, see tables 7.2 and 7.3.
11
0
R
Reserved This bit is always read as 0. The write value should always be 0.
Rev. 6.00 Jun. 12, 2007 Page 118 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Bit 10 9
Bit Name BSZ1 BSZ0
Initial Value 1* 1*
R/W R/W R/W
Description Data Bus Size Specify the data bus width of each area. 00: Reserved (setting prohibited) 01: 8 bits 10: 16 bits 11: Reserved (setting prohibited) Notes: 1. The data bus width for area 0 is specified by the external pin. These bits are ignored. 2. When area 5 or 6 is specified as PCMCIA space, the bus width can be specified as either 8 bits or 16 bits. 3. If area 3 is specified as SDRAM space, the bus width must be specified as 16 bits. 4. These bits must be specified to either 01 or 11 before accessing to memory in other than area 0.
8 to 0
All 0
R
Reserved These bits are always read as 0. The write value should always be 0.
Note:
*
CS0BCR fetches the external pin state (MD3) that specify the bus width at a power-on reset.
Rev. 6.00 Jun. 12, 2007 Page 119 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
7.4.3
CSn Space Wait Control Register (CSnWCR) (n = 0, 3, 4, 5B, 6B)
CSnWCR specifies various wait cycles for memory accesses. The bit configuration of this register varies as shown below according to the memory type (TYPE3, TYPE2, TYPE1, or TYPE0) specified by the CSn space bus control register (CSnBCR). Specify CSnWCR before accessing the target area. Specify CSnBCR first, then specify CSnWCR. Normal Space, Byte-Selection SRAM: * CS0WCR
Bit 31 to 13 Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 12 11 SW1 SW0 0 0 R/W R/W Number of Delay Cycles from Address, CSn Assertion to RD, WEn (BEn) Assertion Specify the number of delay cycles from address and CSn assertion to RD and WEn (BEn) assertion. 00: 0.5 cycles 01: 1.5 cycles 10: 2.5 cycles 11: 3.5 cycles
Rev. 6.00 Jun. 12, 2007 Page 120 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Bit 10 9 8 7
Bit Name WR3 WR2 WR1 WR0
Initial Value 1 0 1 0
R/W R/W R/W R/W R/W
Description Number of Access Wait Cycles Specify the number of wait cycles that are necessary for read or write access. 0000: 0 cycle 0001: 1 cycle 0010: 2 cycles 0011: 3 cycles 0100: 4 cycles 0101: 5 cycles 0110: 6 cycles 0111: 8 cycles 1000: 10 cycles 1001: 12 cycles 1010: 14 cycles 1011: 18 cycles 1100: 24 cycles 1101: Reserved (setting prohibited) 1110: Reserved (setting prohibited) 1111: Reserved (setting prohibited) External Wait Mask Specification Specifies whether or not the external wait input is valid. The specification by this bit is valid even when the number of access wait cycle is 0. 0: External wait is valid 1: External wait is ignored
6
WM
0
R/W
5 to 2
All 0
R
Reserved These bits are always read as 0. The write value should always be 0.
1 0
HW1 HW0
0 0
R/W R/W
Number of Delay Cycles from RD, WEn (BEn) negation to Address, CSn negation Specify the number of delay cycles from RD and WEn (BEn) negation to address and CSn negation. 00: 0.5 cycles 01: 1.5 cycles 10: 2.5 cycles 11: 3.5 cycles
Rev. 6.00 Jun. 12, 2007 Page 121 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
* CS3WCR
Bit 31 to 21 Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 20 BAS 0 R/W Byte Access Selection for Byte-Selection SRAM Specifies the WEn (BEn) and RD/WR signal timing when the byte-selection SRAM interface is used. 0: Asserts the WEn (BEn) signal at the read/write timing (signal used as strobe) and asserts the RD/WR signal during the write access cycle (signal used as status) 1: Asserts the WEn (BEn) signal during the read/write access cycle (used as status) and asserts the RD/WR signal at the write timing (used as strobe) 19 to 11 All 0 R Reserved These bits are always read as 0. The write value should always be 0. 10 9 8 7 WR3 WR2 WR1 WR0 1 0 1 0 R/W R/W R/W R/W Number of Access Wait Cycles Specify the number of wait cycles that are necessary for read access. 0000: 0 cycle 0001: 1 cycle 0010: 2 cycles 0011: 3 cycles 0100: 4 cycles 0101: 5 cycles 0110: 6 cycles 0111: 8 cycles 1000: 10 cycles 1001: 12 cycles 1010: 14 cycles 1011: 18 cycles 1100: 24 cycles 1101: Reserved (setting prohibited) 1110: Reserved (setting prohibited) 1111: Reserved (setting prohibited)
Rev. 6.00 Jun. 12, 2007 Page 122 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Bit 6
Bit Name WM
Initial Value 0
R/W R/W
Description External Wait Mask Specification Specify whether or not the external wait input is valid. The specification by this bit is valid even when the number of access wait cycle is 0. 0: External wait is valid 1: External wait is ignored
5 to 0
All 0
R
Reserved These bits are always read as 0. The write value should always be 0.
* CS4WCR
Bit 31 to 21 Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 20 BAS 0 R/W Byte Access Selection for Byte-Selection SRAM Specifies the WEn (BEn) and RD/WR signal timing when the byte-selection SRAM interface is used. 0: Asserts the WEn (BEn) signal at the read/write timing (signal used as strobe) and asserts the RD/WR signal during the write access cycle (signal used as status) 1: Asserts the WEn (BEn) signal during the read/write access cycle (signal used as status) and asserts the RD/WR signal at the write timing (signal used as strobe) 19 0 R Reserved This bit is always read as 0. The write value should always be 0.
Rev. 6.00 Jun. 12, 2007 Page 123 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Bit 18 17 16
Bit Name WW2 WW1 WW0
Initial Value 0 0 0
R/W R/W R/W R/W
Description Number of Write Access Wait Cycles Specify the number of cycles that are necessary for write access. 000: Same number of cycles as WR3 to WR0 setting (read access wait) 001: 0 cycle 010: 1 cycle 011: 2 cycles 100: 3 cycles 101: 4 cycles 110: 5 cycles 111: 6 cycles
15 to 13
All 0
R
Reserved These bits are always read as 0. The write value should always be 0.
12 11
SW1 SW0
0 0
R/W R/W
Number of Delay Cycles from Address, CSn Assertion to RD, WEn (BEn) Assertion Specify the number of delay cycles from address and CSn assertion to RD and WEn (BEn) assertion. 00: 0.5 cycles 01: 1.5 cycles 10: 2.5 cycles 11: 3.5 cycles
Rev. 6.00 Jun. 12, 2007 Page 124 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Bit 10 9 8 7
Bit Name WR3 WR2 WR1 WR0
Initial Value 1 0 1 0
R/W R/W R/W R/W R/W
Description Number of Access Wait Cycles Specify the number of wait cycles that are necessary for read access. 0000: 0 cycle 0001: 1 cycle 0010: 2 cycles 0011: 3 cycles 0100: 4 cycles 0101: 5 cycles 0110: 6 cycles 0111: 8 cycles 1000: 10 cycles 1001: 12 cycles 1010: 14 cycles 1011: 18 cycles 1100: 24 cycles 1101: Reserved (setting prohibited) 1110: Reserved (setting prohibited) 1111: Reserved (setting prohibited) External Wait Mask Specification Specifies whether or not the external wait input is valid. The specification by this bit is valid even when the number of access wait cycles is 0. 0: External wait is valid 1: External wait is ignored
6
WM
0
R/W
5 to 2
All 0
R
Reserved These bits are always read as 0. The write value should always be 0.
1 0
HW1 HW0
0 0
R/W R/W
Number of Delay Cycles from RD, WEn (BEn) negation to Address, CSn negation Specify the number of delay cycles from RD and WEn (BEn) negation to address and CSn negation. 00: 0.5 cycles 01: 1.5 cycles 10: 2.5 cycles 11: 3.5 cycles
Rev. 6.00 Jun. 12, 2007 Page 125 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
* CS5BWCR
Bit 31 to 19 Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 18 17 16 WW2 WW1 WW0 0 0 0 R/W R/W R/W Number of Write Access Wait Cycles Specify the number of cycles that are necessary for write access. 000: Same number of cycles as WR3 to WR0 setting (read access wait) 001: 0 cycle 010: 1 cycle 011: 2 cycles 100: 3 cycles 101: 4 cycles 110: 5 cycles 111: 6 cycles 15 to 13 All 0 R Reserved These bits are always read as 0. The write value should always be 0. 12 11 SW1 SW0 0 0 R/W R/W Number of Delay Cycles from Address, CSn Assertion to RD, WEn (BEn) Assertion Specify the number of delay cycles from address and CSn assertion to RD and WEn (BEn) assertion. 00: 0.5 cycles 01: 1.5 cycles 10: 2.5 cycles 11: 3.5 cycles
Rev. 6.00 Jun. 12, 2007 Page 126 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Bit 10 9 8 7
Bit Name WR3 WR2 WR1 WR0
Initial Value 1 0 1 0
R/W R/W R/W R/W R/W
Description Number of Access Wait Cycles Specify the number of wait cycles that are necessary for read access. 0000: 0 cycle 0001: 1 cycle 0010: 2 cycles 0011: 3 cycles 0100: 4 cycles 0101: 5 cycles 0110: 6 cycles 0111: 8 cycles 1000: 10 cycles 1001: 12 cycles 1010: 14 cycles 1011: 18 cycles 1100: 24 cycles 1101: Reserved (setting prohibited) 1110: Reserved (setting prohibited) 1111: Reserved (setting prohibited) External Wait Mask Specification Specify whether or not the external wait input is valid. The specification by this bit is valid even when the number of access wait cycle is 0. 0: External wait is valid 1: External wait is ignored
6
WM
0
R/W
5 to 2
All 0
R
Reserved These bits are always read as 0. The write value should always be 0.
1 0
HW1 HW0
0 0
R/W R/W
Number of Delay Cycles from RD, WEn (BEn) negation to Address, CSn negation Specify the number of delay cycles from RD and WEn (BEn) negation to address and CSn negation. 00: 0.5 cycles 01: 1.5 cycles 10: 2.5 cycles 11: 3.5 cycles
Rev. 6.00 Jun. 12, 2007 Page 127 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
* CS6BWCR
Bit 31 to 21 Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 20 BAS 0 R/W Byte Access Selection for Byte-Selection SRAM Specifies the WEn (BEn) and RD/WR signal timing when the byte-selection SRAM interface is used. 0: Asserts the WEn (BEn) signal at the read/write timing (signal used as strobe) and asserts the RD/WR signal during the write access cycle (signal used as status) 1: Asserts the WEn (BEn) signal during the read/write access cycle (used as status) and asserts the RD/WR signal at the write timing (used as strobe) 19 to 13 All 0 R Reserved These bits are always read as 0. The write value should always be 0. 12 11 SW1 SW0 0 0 R/W R/W Number of Delay Cycles from Address, CSn Assertion to RD, WEn (BEn) Assertion Specify the number of delay cycles from address and CSn assertion to RD and WEn (BEn) assertion. 00: 0.5 cycles 01: 1.5 cycles 10: 2.5 cycles 11: 3.5 cycles
Rev. 6.00 Jun. 12, 2007 Page 128 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Bit 10 9 8 7
Bit Name WR3 WR2 WR1 WR0
Initial Value 1 0 1 0
R/W R/W R/W R/W R/W
Description Number of Access Wait Cycles Specify the number of wait cycles that are necessary for read or write access. 0000: 0 cycle 0001: 1 cycle 0010: 2 cycles 0011: 3 cycles 0100: 4 cycles 0101: 5 cycles 0110: 6 cycles 0111: 8 cycles 1000: 10 cycles 1001: 12 cycles 1010: 14 cycles 1011: 18 cycles 1100: 24 cycles 1101: Reserved (setting prohibited) 1110: Reserved (setting prohibited) 1111: Reserved (setting prohibited) External Wait Mask Specification Specifies whether or not the external wait input is valid. The specification by this bit is valid even when the number of access wait cycle is 0. 0: External wait is valid 1: External wait is ignored
6
WM
0
R/W
5 to 2
All 0
R
Reserved These bits are always read as 0. The write value should always be 0.
1 0
HW1 HW0
0 0
R/W R/W
Number of Delay Cycles from RD, WEn (BEn) negation to Address, CSn negation Specify the number of delay cycles from RD and WEn (BEn) negation to address and CSn negation. 00: 0.5 cycles 01: 1.5 cycles 10: 2.5 cycles 11: 3.5 cycles
Rev. 6.00 Jun. 12, 2007 Page 129 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
SDRAM: * CS3WCR
Bit 31 to 15 Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 14 13 WTRP1 WTRP0 0 0 R/W R/W Wait Cycle Number for Precharge Completion Specify the number of minimum wait cycles inserted to wait for the completion of precharge in the following cases. * * * * From the start of auto-precharge to the issuing of the ACTV command for the same bank. From the issuing of the PRE/PALL command to the issuing of the ACTV command for the same bank. From the issuing of the PALL command during auto-refreshing to the issuing of the REF command. From the issuing of the PALL command during self-refreshing to the issuing of the SELF command. 00: 0 cycle (no wait cycle) 01: 1 cycle 10: 2 cycles 11: 3 cycles 12 0 R Reserved This bit is always read as 0. The write value should always be 0. 11 10 WTRCD1 WTRCD0 0 1 R/W R/W Wait Cycle Number from ACTV Command to READ(A)/WRIT(A) Command Specify the number of minimum wait cycles from issuing the ACTV command to issuing the READ(A)/WRIT(A) command. 00: 0 cycle (no wait cycle) 01: 1 cycle 10: 2 cycles 11: 3 cycles
Rev. 6.00 Jun. 12, 2007 Page 130 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Bit 9
Bit Name
Initial Value 0
R/W R
Description Reserved This bit is always read as 0. The write value should always be 0.
8 7
A3CL1 A3CL0
1 0
R/W R/W
CAS Latency for Area 3. Specify the CAS latency for area 3. 00: 1 cycle 01: 2 cycles 10: 3 cycles 11: Reserved (setting prohibited)
6, 5
All 0
R
Reserved These bits are always read as 0. The write value should always be 0.
4 3
TRWL1 TRWL0
0 0
R/W R/W
Wait Cycle Number for Precharge Start Wait Specify the number of minimum wait cycles inserted to wait for the start of precharge in the following cases. * From the issuing of the WRITA command by this LSI to the start of the auto-precharge in the SDRAM. The ACTV command for the same bank is issued after issuing the WRITA command in non-bank active mode. To confirm how many cycles should be needed in the SDRAM between receiving the WRITA command and the auto-precharge start, refer to the data sheets for each SDRAM. Set this bit so that the cycle number in that data sheets should not exceed the cycle number set by this bit. * From the issuing of the WRIT command by this LSI to the issuing of the PRE command. A different row address in the same bank is accessed in bank active mode. 00: 0 cycle (no wait cycle) 01: 1 cycle 10: 2 cycles 11: 3 cycles
Rev. 6.00 Jun. 12, 2007 Page 131 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Bit 2
Bit Name
Initial Value 0
R/W R
Description Reserved This bit is always read as 0. The write value should always be 0.
1 0
WTRC1 WTRC0
0 0
R/W R/W
Idle Cycle Number from REF Command/Self-Refreshing Release to ACTV/REF/MRS Command Specify the number of minimum idle cycles in the following cases. * * From the issuing of the REF command to the issuing of the ACTV/REF/MRS command. From the self-refreshing release to the issuing of the ACTV/REF/MRS command. 00: 2 cycles 01: 3 cycles 10: 5 cycles 11: 8 cycles
PCMCIA: * CS5BWCR, CS6BWCR
Bit 31 to 22 Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 21 20 SA1 SA0 0 0 R/W R/W Space Attribute Specification Specify memory card interface or I/O card interface when the PCMCIA interface is selected. * SA1 0: Specifies memory card interface when A25 = 1 1: Specifies I/O card interface when A25 = 1 * SA0 0: Specifies memory card interface when A25 = 0 1: Specifies I/O card interface when A25 = 0
Rev. 6.00 Jun. 12, 2007 Page 132 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Bit 19 to 15
Bit Name
Initial Value All 0
R/W R
Description Reserved These bits are always read as 0. The write value should always be 0.
14 13 12 11
TED3 TED2 TED1 TED0
0 0 0 0
R/W R/W R/W R/W
Delay from Address to RD or WE Assert Specify the delay time from address output to RD or WE assertion in PCMCIA interface. 0000: 0.5 cycles 0001: 1.5 cycles 0010: 2.5 cycles 0011: 3.5 cycles 0100: 4.5 cycles 0101: 5.5 cycles 0110: 6.5 cycles 0111: 7.5 cycles 1000: Reserved (setting prohibited) 1001: Reserved (setting prohibited) 1010: Reserved (setting prohibited) 1011: Reserved (setting prohibited) 1100: Reserved (setting prohibited) 1101: Reserved (setting prohibited) 1110: Reserved (setting prohibited) 1111: Reserved (setting prohibited)
Rev. 6.00 Jun. 12, 2007 Page 133 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Bit 10 9 8 7
Bit Name PCW3 PCW2 PCW1 PCW0
Initial Value 1 0 1 0
R/W R/W R/W R/W R/W
Description Number of Access Wait Cycles Specify the number of wait cycles to be inserted. 0000: 3 cycles 0001: 6 cycles 0010: 9 cycles 0011: 12 cycles 0100: 15 cycles 0101: 18 cycles 0110: 22 cycles 0111: 26 cycles 1000: 30 cycles 1001: 33 cycles 1010: 36 cycles 1011: 38 cycles 1100: 52 cycles 1101: 60 cycles 1110: 64 cycles 1111: 80 cycles External Wait Mask Specification Specify whether or not the external wait input is valid. The specification by this bit is valid even when the number of access wait cycle is 0. 0: External wait is valid 1: External wait is ignored
6
WM
0
R/W
5, 4
All 0
R
Reserved These bits are always read as 0. The write value should always be 0.
Rev. 6.00 Jun. 12, 2007 Page 134 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Bit 3 2 1 0
Bit Name TEH3 TEH2 TEH1 TEH0
Initial Value 0 0 0 0
R/W R/W R/W R/W R/W
Description Delay from RD or WE Negate to Address Specify the address hold time from RD or WE negation in the PCMCIA interface. 0000: 0.5 cycle 0001: 1.5 cycles 0010: 2.5 cycles 0011: 3.5 cycles 0100: 4.5 cycles 0101: 5.5 cycles 0110: 6.5 cycles 0111: 7.5 cycles 1000: 8.5 cycles 1001: 9.5 cycles 1010: 10.5 cycles 1011: 11.5 cycles 1100: 12.5 cycles 1101: 13.5 cycles 1110: 14.5 cycles 1111: 15.5 cycles
Rev. 6.00 Jun. 12, 2007 Page 135 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
7.4.4
SDRAM Control Register (SDCR)
SDCR specifies the method to refresh and access SDRAM, and the types of SDRAMs to be connected.
Bit 31 to 12 Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 11 RFSH 0 R/W Refresh Control Specifies whether or not the refreshing SDRAM is performed. 0: Refreshing is not performed 1: Refreshing is performed 10 RMODE 0 R/W Refresh Control Specifies whether to perform auto-refreshing or self-refreshing when the RFSH bit is 1. When the RFSH bit is 1 and this bit is 1, self-refreshing starts immediately. When the RFSH bit is 1 and this bit is 0, auto-refreshing starts according to the contents that are set in RTCSR, RTCNT, and RTCOR. 0: Auto-refreshing is performed 1: Self-refreshing is performed 9 0 R Reserved This bit is always read as 0. The write value should always be 0. 8 BACTV 0 R/W Bank Active Mode Specifies whether to access in auto-precharge mode (using READA and WRITA commands) or in bank active mode (using READ and WRIT commands). 0: Auto-precharge mode (using READA and WRITA commands) 1: Bank active mode (using READ and WRIT commands) 7 to 5 All 0 R Reserved These bits are always read as 0. The write value should always be 0.
Rev. 6.00 Jun. 12, 2007 Page 136 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Bit 4 3
Bit Name A3ROW1 A3ROW0
Initial Value 0 0
R/W R/W R/W
Description Number of Bits of Row Address for Area 3 Specify the number of bits of the row address for area 3. 00: 11 bits 01: 12 bits 10: 13 bits 11: Reserved (setting prohibited)
2
0
R
Reserved This bit is always read as 0. The write value should always be 0.
1 0
A3COL1 A3COL0
0 0
R/W R/W
Number of Bits of Column Address for Area 3 Specify the number of bits of the column address for area 3. 00: 8 bits 01: 9 bits 10: 10 bits 11: Reserved (setting prohibited)
7.4.5
Refresh Timer Control/Status Register (RTCSR)
RTCSR specifies various items about refresh for SDRAM. When RTCSR is written to, the upper 16 bits of the write data must be H'A55A to cancel write protection.
Bit 31 to 8 Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0.
Rev. 6.00 Jun. 12, 2007 Page 137 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Bit 7
Bit Name CMF
Initial Value 0
R/W R/W
Description Compare Match Flag Indicates that a compare match occurs between the refresh timer counter (RTCNT) and refresh time constant register (RTCOR). [Clearing condition] When 0 is written to this bit after reading RTCSR with CMF = 1. [Setting condition] When RTCNT value matches RTCOR value
6
0
R
Reserved This bit is always read as 0. The write value should always be 0.
5 4 3
CKS2 CKS1 CKS0
0 0 0
R/W R/W R/W
Clock Select Select the clock input to count-up the refresh timer counter (RTCNT). 000: Stop the counting-up 001: B/4 010: B/16 011: B/64 100: B/256 101: B/1024 110: B/2048 111: B/4096
Rev. 6.00 Jun. 12, 2007 Page 138 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Bit 2 1 0
Bit Name RRC2 RRC1 RRC0
Initial Value 0 0 0
R/W R/W R/W R/W
Description Refresh Count Specify the number of consecutive refresh cycles, when the refresh request occurs after the coincidence of the values of the refresh timer counter (RTCNT) and the refresh time constant register (RTCOR). Using consecutive refresh cycles can prolong cycles between refreshing. 000: Once 001: Twice 010: 4 times 011: 6 times 100: 8 times 101: Reserved (setting prohibited) 110: Reserved (setting prohibited) 111: Reserved (setting prohibited)
7.4.6
Refresh Timer Counter (RTCNT)
RTCNT is an 8-bit counter that increments using the clock selected by bits CKS2 to CKS0 in RTCSR. When RTCNT matches RTCOR, RTCNT is cleared to 0. The value in RTCNT returns to 0 after counting up to 255. When RTCNT is written to, the upper 16 bits of the write data must be H'A55A to cancel write protection.
Bit 31 to 8 Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 7 to 0 All 0 R/W 8-bit Counter
Rev. 6.00 Jun. 12, 2007 Page 139 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
7.4.7
Refresh Time Constant Register (RTCOR)
RTCOR is an 8-bit register. When RTCOR matches RTCNT, the CMF bit in RTCSR is set to 1 and RTCNT is cleared to 0. When the RFSH bit in SDCR is 1, a memory refresh request is issued. The request is maintained until the refresh operation is performed. If the request is not processed when the next matching occurs, the previous request is ignored. When the RTCOR is written to, the upper 16 bits of the write data must be H'A55A to cancel write protection.
Bit 31 to 8 Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 7 to 0 All 0 R/W 8-bit Counter
Rev. 6.00 Jun. 12, 2007 Page 140 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
7.5
7.5.1
Operation
Endian/Access Size and Data Alignment
This LSI supports big endian, in which the most significant byte (MSByte) of multiple byte data is stored in the lower address, and little endian, in which the least significant byte (LSByte) of multiple byte data is stored in the lower address. Endian is specified at a power-on reset by the external pin (MD5). When pin MD5 is driven low at a power-on reset, the endian will become big endian and when pin MD5 is driven high at a power-on reset, the endian will become little endian. Two data bus widths (8 bits and 16 bits) are available for normal memory and byte-selection SRAM. Only 16-bit data bus width is available for SDRAM. Two data bus widths (8 bits and 16 bits) are available for PCMCIA interface. Data alignment is performed in accordance with the data bus width of the device and endian. This also means that when longword data is read from a byte-width device, the read operation must be done four times. In this LSI, data alignment and conversion of data length is performed automatically between the respective interfaces.
Rev. 6.00 Jun. 12, 2007 Page 141 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Tables 7.6 to 7.9 show the relationship between endian, device data width, and access unit. Table 7.6 16-Bit External Device/Big Endian Access and Data Alignment
Data Bus Operation Byte access at 0 Byte access at 1 Byte access at 2 Byte access at 3 Word access at 0 Word access at 2 Longword access at 0 1st time at 0 Strobe Signals
WE3(BE3), WE2(BE2), WE1(BE1), WE0(BE0), D31 to D24 D23 to D16 D15 to D8 D7 to D0 DQMUU DQMUL DQMLU DQMLL

Data 7 to 0 Data 7 to 0 Data 15 to 8 Data 15 to 8 Data 31 to 24 Data 15 to 8
Data 7 to 0 Data 7 to 0 Data 7 to 0 Data 15 to 8 Data 23 to 16 Data 7 to 0

Assert Assert Assert Assert Assert Assert
Assert Assert Assert Assert Assert Assert
2nd time at 2
Rev. 6.00 Jun. 12, 2007 Page 142 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Table 7.7
8-Bit External Device/Big Endian Access and Data Alignment
Data Bus Strobe Signals
WE3(BE3), WE2(BE2), WE1(BE1), WE0(BE0),
Operation Byte access at 0 Byte access at 1 Byte access at 2 Byte access at 3 Word access at 0 1st time at 0
D31 to D24 D23 to D16 D15 to D8
D7 to D0
DQMUU
DQMUL
DQMLU
DQMLL

Data 7 to 0 Data 7 to 0 Data 7 to 0 Data 7 to 0 Data 15 to 8 Data 7 to 0 Data 15 to 8 Data 7 to 0 Data 31 to 24 Data 23 to 16 Data 15 to 8 Data 7 to 0

Assert Assert Assert Assert Assert Assert Assert Assert Assert Assert Assert Assert
2nd time at 1 Word access at 2 1st time at 2
2nd time at 3 Longword access at 0 1st time at 0
2nd time at 1 3rd time at 2 4th time at 3
Rev. 6.00 Jun. 12, 2007 Page 143 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Table 7.8
16-Bit External Device/Little Endian Access and Data Alignment
Data Bus Strobe Signals
WE3(BE3), WE2(BE2), WE1(BE1), WE0(BE0),
Operation Byte access at 0 Byte access at 1 Byte access at 2 Byte access at 3 Word access at 0 Word access at 2 Longword access at 0 1st time at 0
D31 to D24 D23 to D16 D15 to D8
D7 to D0
DQMUU
DQMUL
DQMLU
DQMLL

Data 7 to 0 Data 7 to 0 Data 15 to 8 Data 15 to 8 Data 15 to 8 Data 31 to 24
Data 7 to 0 Data 7 to 0 Data 7 to 0 Data 7 to 0 Data 7 to 0 Data 23 to 16

Assert Assert Assert Assert Assert Assert
Assert Assert Assert Assert Assert Assert
2nd time at 2
Rev. 6.00 Jun. 12, 2007 Page 144 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Table 7.9
8-Bit External Device/Little Endian Access and Data Alignment
Data Bus Strobe Signals
WE3(BE3), WE2(BE2), WE1(BE1), WE0(BE0),
Operation Byte access at 0 Byte access at 1 Byte access at 2 Byte access at 3 Word access at 0 1st time at 0
D31 to D24 D23 to D16 D15 to D8 D7 to D0
DQMUU
DQMUL
DQMLU
DQMLL

Data 7 to 0 Data 7 to 0 Data 7 to 0 Data 7 to 0 Data 7 to 0 Data 15 to 8 Data 7 to 0 Data 15 to 8 Data 7 to 0 Data 15 to 8 Data 23 to 16 Data 31 to 24

Assert Assert Assert Assert Assert Assert Assert Assert Assert Assert Assert Assert
2nd time at 1 Word access at 2 1st time at 2
2nd time at 3 Longword access at 0 1st time at 0
2nd time at 1 3rd time at 2 4th time at 3
Rev. 6.00 Jun. 12, 2007 Page 145 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
7.5.2
Normal Space Interface
Basic Timing: For access to a normal space, this LSI uses strobe signal output in consideration of the fact that mainly static RAM will be directly connected. When using SRAM with a byte-selection pin, see section 7.5.6, Byte-Selection SRAM Interface. Figure 7.3 shows the basic timings of normal space access. A no-wait normal access is completed in two cycles. The BS signal is asserted for one cycle to indicate the start of a bus cycle.
T1 CKIO T2
A
CSn
RD/WR
Read
RD D
RD/WR
WEn(BEn) Write D
BS
Figure 7.3 Normal Space Basic Access Timing (No-Wait Access) There is no output signal which informs external devices of the access size when reading. Although the least significant bit of the address indicates the correct address when the access starts, 16-bit data is always read from a 16-bit device. When writing, only the WEn (BEn) signal for the byte to be written to is asserted.
Rev. 6.00 Jun. 12, 2007 Page 146 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
When buffers are placed on the data bus, the RD signal should be used to control the buffers. The RD/WR signal indicates the same state as a read cycle (driven high) when no access has been carried out. Therefore, care must be taken when controlling the buffers with the RD/WR signal, to avoid data conflict. Figures 7.4 and 7.5 show the basic timings of normal space consecutive access. If the WM bit in CSnWCR is cleared to 0, a Tnop cycle is inserted to check the external wait (figure 7.4). If the WM bit in CSnWCR is set to 1, an external wait request is ignored and no Tnop cycle is inserted (figure 7.5).
T1 CKIO
T2
Tnop
T1
T2
A25 to A0
CSn
RD/WR
RD Read D15 to D0
WEn(BEn) Write D15 to D0
BS
WAIT
Figure 7.4 Consecutive Access to Normal Space (1): Bus Width = 16 bits, Longword Access, CSnWCR.WM = 0 (Access Wait = 0, Cycle Wait = 0)
Rev. 6.00 Jun. 12, 2007 Page 147 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
T1
T2
T1
T2
CKIO
A25 to A0
CSn
RD/WR
RD Read D15 to D0 WEn(BEn) Write D15 to D0
BS
WAIT
Figure 7.5 Consecutive Access to Normal Space (2): Bus Width = 16 bits, Longword Access, CSnWCR.WM = 1 (Access Wait = 0, Cycle Wait = 0)
Rev. 6.00 Jun. 12, 2007 Page 148 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
This LSI
**** ****
128 kwords x 8 bits SRAM
**** **** **** ****
A17 A1 CSn RD D15
****
A16 A0 CS OE I/O7 I/O0 WE
****
****
D0 WE0(BE0)
****
D8 WE1(BE1) D7
****
****
****
A16 A0 CS OE I/O7 I/O0 WE
Figure 7.6 Example of 16-Bit Data-Width SRAM Connection
128 kwords x 8 bits SRAM A16
This LSI A16
...
...
****
A0 CSn RD D7 D0 WE0(BE0)
A0 CS OE I/O7
...
...
I/O0 WE
Figure 7.7 Example of 8-Bit Data-Width SRAM Connection
Rev. 6.00 Jun. 12, 2007 Page 149 of 610 REJ09B0131-0600
...
...
Section 7 Bus State Controller (BSC)
7.5.3
Access Wait Control
Wait cycle insertion on a normal space access can be controlled by the settings of bits WR3 to WR0 in CSnWCR. It is possible for areas 4, 5A, and 5B to insert wait cycles independently in read access and in write access. The areas other than 4, 5A, and 5B have the same access wait for read cycle and write cycle. The specified number of Tw cycles is inserted as wait cycles in a normal space access shown in figure 7.9.
T1 Tw T2
CKIO A25 to A0 CSn RD/WR RD Read D15 to D0 WEn(BEn) Write D15 to D0 BS
Figure 7.8 Wait Timing for Normal Space Access (Software Wait Only) When the WM bit in CSnWCR is cleared to 0, the external wait signal (WAIT) is also sampled. The WAIT pin sampling is shown in figure 7.9. In this example, two wait cycles are inserted as software wait. The WAIT signal is sampled at the falling edge of the CKIO signal in the cycle immediately before the T2 cycle (T1 or Tw cycle).
Rev. 6.00 Jun. 12, 2007 Page 150 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
T1 CKIO A25 to A0 CSn RD/WR RD Read D15 to D0 WEn(BEn) Write D15 to D0 WAIT BS
Tw
Tw
Wait cycles inserted by WAIT signal Twx T2
Figure 7.9 Wait Cycle Timing for Normal Space Access (Wait cycle Insertion using WAIT)
Rev. 6.00 Jun. 12, 2007 Page 151 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
7.5.4
Extension of Chip Select (CSn) Assertion Period
The number of cycles from CSn assertion to RD and WEn (BEn) assertion can be specified by setting bits SW1 and SW0 in CSnWCR. The number of cycles from RD and WEn (BEn) negation to CSn negation can be specified by setting bits HW1 and HW0. Therefore, a flexible interface to an external device can be obtained. Figure 7.10 shows an example. A Th cycle and a Tf cycle are added before and after a normal cycle, respectively. In these cycles, RD and WEn (BEn) are not asserted, while other signals are asserted. The data output is prolonged to the Tf cycle, and this prolongation is useful for devices with slow writing operations.
Th T1 T2 Tf
CKIO A25 to A0 CSn RD/WR RD Read D15 to D0 WEn(BEn) Write D15 to D0 BS
Figure 7.10 Example of Timing when CSn Assertion Period is Extended
Rev. 6.00 Jun. 12, 2007 Page 152 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
7.5.5
SDRAM Interface
SDRAM Direct Connection: The SDRAM that can be connected to this LSI is a product that has 11/12/13 bits of row address, 8/9/10 bits of column address, 4 or less banks, and uses the A10 pin for setting precharge mode in read and write command cycles. The control signals for direct connection of SDRAM are RAS, CAS, RD/WR, DQMLU, DQMLL, CKE, and CS3. Signals other than CKE are valid when CS3 is asserted. SDRAM can be connected to area 2. The data bus width of the area that is connected to SDRAM can be set to 16 bits. Burst read/single write (burst length 1) and burst read/burst write (burst length 1) are supported as the SDRAM operating mode. Commands for SDRAM can be specified by RAS, CAS, RD/WR, and specific address signals. These commands are shown below. * * * * * * * * * * * NOP Auto-refreshing (REF) Self-refreshing (SELF) All banks precharge (PALL) Specified bank precharge (PRE) Bank active (ACTV) Read (READ) Read with precharge (READA) Write (WRIT) Write with precharge (WRITA) Write mode register (MRS)
The byte to be accessed is specified by DQMLU and DQMLL. Reading or writing is performed for a byte whose corresponding DQMxx is low. For details on the relationship between DQMxx and the byte to be accessed, refer to section 7.5.1, Endian/Access Size and Data Alignment.
Rev. 6.00 Jun. 12, 2007 Page 153 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Figure 7.11 shows an example of the connection of the SDRAM with the LSI.
This LSI
A14
64-Mbit SDRAM (1 Mword x 16 bits x 4 banks) A13 A0 CKE CLK CS
A1 CKE CKIO CSn
...
RAS CAS RD/WR D15
RAS CAS WE I/O15 I/O0 DQMU DQML
D0 DQMLU DQMLL
Figure 7.11 Example of 16-Bit Data-Width SDRAM Connection Address Multiplexing: An address multiplexing is specified so that SDRAM can be connected without external multiplexing circuitry according to the setting of bits BSZ1 and BSZ0 in CSnBCR, AnROW1 and AnROW0 and AnCOL1 AnCOL0 in SDCR. Tables 7.10 to 7.12 show the relationship between those settings and the bits output on the address pins. Do not specify those bits in the manner other than this table, otherwise the operation of this LSI is not guaranteed. A25 to A18 are not multiplexed and the original values of address are always output on these pins. Pin A0 of SDRAM specifies a word address. Therefore, connect this A0 pin of SDRAM to pin A1 of this LSI; pin A1 pin of SDRAM to pin A2 of this LSI, and so on.
Rev. 6.00 Jun. 12, 2007 Page 154 of 610 REJ09B0131-0600
...
...
...
Section 7 Bus State Controller (BSC)
Table 7.10 Relationship between Register Settings and Address Multiplex Output (1) Conditions: One 16-Mbit product (512 kwords x 16 bits x 2 banks, 8-bit column product) is connected with A3BSZ[1:0] = 10 (16-bit data bus width), A3ROW[1:0] = 00 (11-bit row address), and A3COL[1:0] = 00 (8-bit column address).
Output Row Pins of this LSI Address A17 A16 A15 A14 A13 A12 A11 A10 A9 A8 A7 A6 A5 A4 A3 A2 A1 A0 A25 A24 A23 A22 A21 A20*2 A19 A18 A17 A16 A15 A14 A13 A12 A11 A10 A9 A8 Output Column Address Pins of SDRAM Function A17 A16 A15 A14 A21 A20*2 L/H* A10 A9 A8 A7 A6 A5 A4 A3 A2 A1 A0
1
Unused
A11 (BA0) A10/AP A9 A8 A7 A6 A5 A4 A3 A2 A1 A0
Specifies bank Specifies address/precharge Address
Unused
Notes: 1. L/H is a bit used in the command specification; it is fixed low or high according to the access mode. 2. Bank address specification
Rev. 6.00 Jun. 12, 2007 Page 155 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Table 7.11 Relationship between Register Settings and Address Multiplex Output (2) Conditions: One 64-Mbit product (1 Mword x 16 bits x 4 banks, 8-bit column product) is connected with A3BSZ[1:0] = 10 (16-bit data bus width), A3ROW[1:0] = 01 (12-bit row address), and A3COL[1:0] = 00 (8-bit column address).
Output Row Pins of this LSI Address A17 A16 A15 A14 A13 A12 A11 A10 A9 A8 A7 A6 A5 A4 A3 A2 A1 A0 A25 A24 A23 A22* A21* A20 A19 A18 A17 A16 A15 A14 A13 A12 A11 A10 A9 A8
2 2
Output Column Address Pins of SDRAM Function A17 A16 A15 A22*2 A21* A12 L/H* A10 A9 A8 A7 A6 A5 A4 A3 A2 A1 A0
1 2
Unused
A13 (BA1) A12 (BA0) A11 A10/AP A9 A8 A7 A6 A5 A4 A3 A2 A1 A0
Specifies bank
Address Specifies address/precharge Address
Unused
Notes: 1. L/H is a bit used in the command specification; it is fixed low or high according to the access mode. 2. Bank address specification
Rev. 6.00 Jun. 12, 2007 Page 156 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Table 7.12 Relationship between Register Settings and Address Multiplex Output (3) Conditions: One 128-Mbit product (2 Mwords x 16 bits x 4 banks, 9-bit column product) is connected with A3BSZ[1:0] = 10 (16-bit data bus width), A3ROW[1:0] = 01 (12-bit row address), and A3COL[1:0] = 01 (9-bit column address).
Output Row Pins of this LSI Address A17 A16 A15 A14 A13 A12 A11 A10 A9 A8 A7 A6 A5 A4 A3 A2 A1 A0 A26 A25 A24 A23* A22* A21 A20 A19 A18 A17 A16 A15 A14 A13 A12 A11 A10 A9
2 2
Output Column Address Pins of SDRAM Function A17 A16 A15 A23*2 A22* A12 L/H* A10 A9 A8 A7 A6 A5 A4 A3 A2 A1 A0
1 2
Unused
A13 (BA1) A12 (BA0) A11 A10/AP A9 A8 A7 A6 A5 A4 A3 A2 A1 A0
Specifies bank
Address Specifies address/precharge Address
Unused
Notes: 1. L/H is a bit used in the command specification; it is fixed low or high according to the access mode. 2. Bank address specification
Rev. 6.00 Jun. 12, 2007 Page 157 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Table 7.13 Relationship between Register Settings and Address Multiplex Output (4) Conditions: One 256-Mbit product (4 Mwords x 16 bits x 4 banks, 10-bit column product) is connected with A3BSZ[1:0] = 10 (16-bit data bus width), A3ROW[1:0] = 01 (12-bit row address), and A3COL[1:0] = 10 (10-bit column address).
Output Row Pins of this LSI Address A17 A16 A15 A14 A13 A12 A11 A10 A9 A8 A7 A6 A5 A4 A3 A2 A1 A0 A27 A26 A25 A24* A23* A22 A21 A20 A19 A18 A17 A16 A15 A14 A13 A12 A11 A10
2 2
Output Column Address Pins of SDRAM Function A17 A16 A15 A24*2 A23* A12 L/H* A10 A9 A8 A7 A6 A5 A4 A3 A2 A1 A0
1 2
Unused
A13 (BA1) A12 (BA0) A11 A10/AP A9 A8 A7 A6 A5 A4 A3 A2 A1 A0
Specifies bank
Address Specifies address/precharge Address
Unused
Notes: 1. L/H is a bit used in the command specification; it is fixed low or high according to the access mode. 2. Bank address specification
Rev. 6.00 Jun. 12, 2007 Page 158 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Table 7.14 Relationship between Register Settings and Address Multiplex Output (5) Conditions: One 256-Mbit product (4 Mwords x 16 bits x 4 banks, 9-bit column product) is connected with A3BSZ[1:0] = 10 (16-bit data bus width), A3ROW[1:0] = 10 (13-bit row address), and A3COL[1:0] = 01 (9-bit column address).
Output Row Pins of this LSI Address A17 A16 A15 A14 A13 A12 A11 A10 A9 A8 A7 A6 A5 A4 A3 A2 A1 A0 A26 A25 A24* A23* A22 A21 A20 A19 A18 A17 A16 A15 A14 A13 A12 A11 A10 A9
2 2
Output Column Address Pins of SDRAM Function A17 A16 A24*2 A23* A13 A12 L/H*1 A10 A9 A8 A7 A6 A5 A4 A3 A2 A1 A0
2
Unused
A14 (BA1) A13 (BA0) A12 A11 A10/AP A9 A8 A7 A6 A5 A4 A3 A2 A1 A0
Specifies bank
Address
Specifies address/precharge Address
Unused
Notes: 1. L/H is a bit used in the command specification; it is fixed low or high according to the access mode. 2. Bank address specification
Rev. 6.00 Jun. 12, 2007 Page 159 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Table 7.15 Relationship between Register Settings and Address Multiplex Output (6) Conditions: One 512-Mbit product (8 Mwords x 16 bits x 4 banks, 10-bit column product) is connected with A3BSZ[1:0] = 10 (16-bit data bus width), A3ROW[1:0] = 10 (13-bit row address), and A3COL[1:0] = 10 (10-bit column address).
Output Row Pins of this LSI Address A17 A16 A15 A14 A13 A12 A11 A10 A9 A8 A7 A6 A5 A4 A3 A2 A1 A0 A27 A26 A25* A24* A23 A22 A21 A20 A19 A18 A17 A16 A15 A14 A13 A12 A11 A10
2 2
Output Column Address Pins of SDRAM Function A17 A16 A25*2 A24* A13 A12 L/H*1 A10 A9 A8 A7 A6 A5 A4 A3 A2 A1 A0
2
Unused
A14 (BA1) A13 (BA0) A12 A11 A10/AP A9 A8 A7 A6 A5 A4 A3 A2 A1 A0
Specifies bank
Address
Specifies address/precharge Address
Unused
Notes: 1. L/H is a bit used in the command specification; it is fixed low or high according to the access mode. 2. Bank address specification
Rev. 6.00 Jun. 12, 2007 Page 160 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Burst Read: A burst read occurs in the following cases with this LSI. 1. Access size in reading is larger than data bus width. 2. 16-byte transfer in cache miss. 3. 16-byte transfer by E-DMAC (access to non-cacheable area) This LSI always accesses the SDRAM with burst length 1. For example, read access of burst length 1 is performed consecutively eight times to read 16-byte consecutive data from the SDRAM that is connected to a 16-bit data bus. Table 7.16 shows the relationship between the access size and the number of bursts. Table 7.16 Relationship between Access Size and Number of Bursts
Bus Width 16 bits Access Size 8 bits 16 bits 32 bits 16 bytes Number of Bursts 1 1 2 8
Figures 7.12 and 7.13 show timing charts in burst read. In burst read, the ACTV command is output in the Tr cycle, the READ command is issued in the Tc1, Tc2, and Tc3 cycles, the READA command is issued in the Tc4 cycle, and the read data is latched at the rising edge of the external clock (CKIO) in the Td1 to Td4 cycles. The Tap cycle is used to wait for the completion of an auto-precharge induced by the READ command in the SDRAM. In the Tap cycle, a new command will not be issued to the same bank. However, other banks can be accessed. The number of Tap cycles is specified by bits WTRP1 and WTRP0 in CS3WCR. In this LSI, wait cycles can be inserted by specifying bits in CSnWCR to connect the SDRAM with variable frequencies. Figure 7.15 shows an example in which wait cycles are inserted. The number of cycles from the Tr cycle where the ACTV command is output to the Tc1 cycle where the READA command is output can be specified using bits WTRCD1 and WTRCD0 in CS3WCR. When bits WTRCD1 and WTRCD0 is set to one cycle or more, a Trw cycle where the NOP command is issued is inserted between the Tr cycle and Tc1 cycle. The number of cycles from the Tc1 cycle where the READA command is output to the Td1 cycle where the read data is latched can be specified by bits A3CL1 and A3CL0 bits in CS3WCR in CS3WCR. This number of cycles corresponds to the synchronous DRAM CAS latency. The CAS latency for the synchronous DRAM is normally defined as up to three cycles. However, the CAS latency in this LSI can be specified as one to four cycles. This CAS latency can be achieved by connecting a latch circuit between this LSI and the synchronous DRAM.
Rev. 6.00 Jun. 12, 2007 Page 161 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Tr CKIO A25 to A0 A11* CSn RAS CAS RD/WR DQMxx D15 to D0 BS
Tc1
Td1 Tc2
Td2 Tc3
Td3 Tc4
Td4 Tde Tap
Note: * Address pin to be connected to pin A10 of SDRAM.
Figure 7.12 Burst Read Basic Timing (Auto Precharge)
Rev. 6.00 Jun. 12, 2007 Page 162 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Tr CKIO A25 to A0 A11* CSn RAS CAS RD/WR DQMxx D15 to D0 BS
Trw
Tc1
Tw Tc2
Td1 Tc3
Td2 Tc4
Td3
Td4 Tde Tap
Note: * Address pin to be connected to pin A10 of SDRAM.
Figure 7.13 Burst Read Wait Specification Timing (Auto Precharge)
Rev. 6.00 Jun. 12, 2007 Page 163 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Single Read: A read access ends in one cycle when data exists in non-cacheable area and the data bus width is larger than or equal to access size. Since the burst length is set to 1 in synchronous DRAM burst read/single write mode, only the required data is output. Consequently, no unnecessary bus cycles are generated even when a cache-through area is accessed. Figure 7.14 shows the single read basic timing.
Tr CKIO A25 to A0 A11* CSn RAS CAS RD/WR DQMxx D15 to D0 BS
Tc1
Td1
Tde
Tap
Note: * Address pin to be connected to pin A10 of SDRAM.
Figure 7.14 Basic Timing for Single Read (Auto Precharge)
Rev. 6.00 Jun. 12, 2007 Page 164 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Burst Write: A burst write occurs in the following cases in this LSI. 1. Access size in writing is larger than data bus width. 2. Write-back of the cache 3. 16-byte transfer by E-DMAC (access to non-cacheable area) This LSI always accesses SDRAM with burst length 1. For example, write access of burst length 1 is performed consecutively eight times to write 16-byte consecutive data to the SDRAM that is connected to a 16-bit data bus. The relationship between the access size and the number of bursts is shown in table 7.13. Figure 7.15 shows a timing chart for burst writes. In burst write, the ACTV command is output in the Tr cycle, the WRIT command is issued in the Tc1, Tc2, and Tc3 cycles, and the WRITA command is issued to execute an auto-precharge in the Tc4 cycle. In the write cycle, the write data is output simultaneously with the write command. After the write command with the auto-precharge is output, the Trw1 cycle that waits for the auto-precharge initiation is followed by the Tap cycle that waits for completion of the auto-precharge induced by the WRITA command in the SDRAM. In the Tap cycle, a new command will not be issued to the same bank. However, other CS areas and other banks can be accessed. The number of Trw1 cycles is specified by bits TRWL1 and TRWL0 in CS3WCR. The number of Tap cycles is specified by bits WTRP1 and WTRP0 in CS3WCR.
Rev. 6.00 Jun. 12, 2007 Page 165 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Tr CKIO A25 to A0 A11* CSn RAS CAS RD/WR DQMxx D15 to D0 BS
Tc1
Tc2
Tc3
Tc4
Trwl
Tap
Note: * Address pin to be connected to pin A10 of SDRAM.
Figure 7.15 Basic Timing for Burst Write (Auto Precharge)
Rev. 6.00 Jun. 12, 2007 Page 166 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Single Write: A write access ends in one cycle when data is written in non-cacheable area and the data bus width is larger than or equal to access size. Figure 7.16 shows the single write basic timing.
Tr CKIO A25 to A0 A11* CSn RAS CAS RD/WR DQMxx D15 to D0 BS
Tc1
Trwl
Tap
Note: * Address pin to be connected to pin A10 of SDRAM.
Figure 7.16 Basic Timing for Single Write (Auto-Precharge) Bank Active: The synchronous DRAM bank function is used to support high-speed accesses to the same row address. When the BACTV bit in SDCR is 1, accesses are performed using commands without auto-precharge (READ or WRIT). This function is called bank-active function. When a bank-active function is used, precharging is not performed when the access ends. When accessing the same row address in the same bank, it is possible to issue the READ or WRIT command immediately, without issuing an ACTV command. Since synchronous DRAM is internally divided into several banks, it is possible to keep one row address in each bank activated. If the next access is to a different row address, a PRE command is first issued to precharge the relevant bank, then when precharging is completed, the access is performed by issuing an ACTV command followed by a READ or WRIT command. If this is followed by an access to a different row address, the access time will be longer because of the precharging performed after the access request is issued. The number of cycles between issuance of the PRE command and the ACTV command is determined by bits WTRP1 and WTRP0 in CSnWCR.
Rev. 6.00 Jun. 12, 2007 Page 167 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
In a write access, when an auto-precharge is performed, a command cannot be issued to the same bank for a period of Trwl + Tap cycles after issuance of the WRITA command. When bank active mode is used, READ or WRIT command can be issued successively if the row address is the same. The number of cycles can thus be reduced by Trwl + Tap cycles for each write. There is a limit on tRAS, the time for placing each bank in the active state. If there is no guarantee that another row address will be accessed within the period in which this value is maintained by program execution, it is necessary to set auto-refreshing and set the refresh cycle to no more than the maximum value of tRAS. A burst read cycle without auto-precharge is shown in figure 7.17, a burst read cycle for the same row address in figure 7.18, and a burst read cycle for different row addresses in figure 7.19. Similarly, a single write cycle without auto-precharge is shown in figure 7.20, a single write cycle for the same row address in figure 7.21, and a single write cycle for different row addresses in figure 7.21. In figure 7.18, a Tnop cycle in which no operation is performed is inserted before the Tc cycle that issues the READ command. The Tnop cycle is inserted to secure two cycles of CAS latency for the DQMxx signal that specifies which byte data is read from SDRAM. If the CAS latency is specified as two cycles or more, the Tnop cycle is not inserted because the two cycles of latency can be secured even if the DQMxx signal is asserted after the Tc cycle. When bank active mode is set, if only accesses to the respective banks in the area 3 are considered, as long as accesses to the same row address continue, the operation starts with the cycle in figure 7.17 or 7.20, followed by repetition of the cycle in figure 7.18 or 7.21. An access to a different area during this time has no effect. When a different row address is accessed in the bank active state, the bus cycle shown in figure 7.19 or 7.22 is executed instead of that in figure 7.18 or 7.21. In bank active mode, too, all banks become inactive after a refresh cycle.
Rev. 6.00 Jun. 12, 2007 Page 168 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Tr CKIO A25 to A0 A11* CSn RAS CAS RD/WR DQMxx D15 to D0 BS
Tc1
Td1 Tc2
Td2 Tc3
Td3 Tc4
Td4 Tde
Note: * Address pin to be connected to pin A10 of SDRAM.
Figure 7.17 Burst Read Timing (No Auto Precharge)
Rev. 6.00 Jun. 12, 2007 Page 169 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Tnop CKIO A25 to A0 A11* CSn RAS CAS RD/WR DQMxx D15 to D0 BS
Tc1
Td1 Tc2
Td2 Tc3
Td3 Tc4
Td4 Tde
Note: * Address pin to be connected to pin A10 of SDRAM.
Figure 7.18 Burst Read Timing (Bank Active, Same Row Address)
Rev. 6.00 Jun. 12, 2007 Page 170 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Tp CKIO A25 to A0 A11* CSn RAS CAS RD/WR DQMxx D15 to D0 BS
Tpw
Tr
Tc1
Td1 Tc2
Td2 Tc3
Td3 Tc4
Td4 Tde
Note: * Address pin to be connected to pin A10 of SDRAM.
Figure 7.19 Burst Read Timing (Bank Active, Different Row Addresses)
Rev. 6.00 Jun. 12, 2007 Page 171 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Tr CKIO A25 to A0 A11* CSn RAS CAS RD/WR DQMxx D15 to D0 BS
Tc1
Note: * Address pin to be connected to pin A10 of SDRAM.
Figure 7.20 Single Write Timing (No Auto Precharge)
Rev. 6.00 Jun. 12, 2007 Page 172 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Tnop CKIO A25 to A0 A11* CSn RAS CAS RD/WR DQMxx D15 to D0 BS
Tc1
Note: * Address pin to be connected to pin A10 of SDRAM.
Figure 7.21 Single Write Timing (Bank Active, Same Row Address)
Rev. 6.00 Jun. 12, 2007 Page 173 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Tp CKIO A25 to A0 A11* CSn RAS CAS RD/WR DQMxx D15 to D0 BS
Tpw
Tr
Tc1
Note: * Address pin to be connected to pin A10 of SDRAM.
Figure 7.22 Single Write Timing (Bank Active, Different Row Addresses) Refreshing: This LSI has a function for controlling synchronous DRAM refreshing. Auto-refreshing can be performed by clearing the RMODE bit to 0 and setting the RFSH bit to 1 in SDCR. A consecutive refreshing can be performed by setting bits RRC2 to RRC0 in RTCSR. If synchronous DRAM is not accessed for a long period, self-refreshing mode, in which the power consumption for data retention is low, can be activated by setting both the RMODE bit and the RFSH bit to 1. 1. Auto-refreshing Refreshing is performed at intervals determined by the input clock selected by bits CKS2 to CKS0 in RTCSR, and the value set by in RTCOR. The value of bits CKS[2:0] in RTCOR should be set so as to satisfy the given refresh interval for the synchronous DRAM used. First make the settings for RTCOR, RTCNT, and the RMODE, then make the CKS[2:0] and RRC[2:0] settings. When the clock is selected by bits CKS[2:0], RTCNT starts counting up from the value at that time. The RTCNT value is constantly compared with the RTCOR value, and if the two values are the same, a refresh request is generated and an auto-refreshing is performed for the number of times specified by the RRC[2:0]. At the same time, RTCNT is cleared to 0 and the count-up is restarted.
Rev. 6.00 Jun. 12, 2007 Page 174 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Figure 7.23 shows the auto-refreshing cycle timing. After starting the auto-refreshing, PALL command is issued in the Tp cycle to make all the banks to precharged state from active state when some bank is being precharged. Then the REF command is issued in the Trr cycle after inserting idle cycles of which number is specified by bits WTRP1 and WTRP0 in CSnWCR. A new command is not issued for the duration of the number of cycles specified by bits WTRC1 and WTRC0 in CSnWCR after the Trr cycle. Bits WTRC1 and WTRC0 in CSnWCR must be set so as to satisfy the SDRAM refreshing cycle time (tRC). A Tpw cycle is inserted between the Tp cycle and Trr cycle when the setting of bits WTRP1 and WTRP0 in CSnWCR is longer than or equal to one cycle.
Tp CKIO A25 to A0 A11* CSn RAS CAS RD/WR DQMxx D15 to D0 BS
Tpw
Trr
Trc
Trc
Trc
Hi-z
Note: * Address pin to be connected to pin A10 of SDRAM.
Figure 7.23 Auto-Refreshing Timing
Rev. 6.00 Jun. 12, 2007 Page 175 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
2. Self-refreshing When self-refreshing mode is selected, the refresh timing and refresh addresses are generated within the synchronous DRAM. Self-refreshing is activated by setting both the RMODE bit and the RFSH bit in SDCR to 1. After starting the self-refreshing, the PALL command is issued in the Tp cycle after the completion of pre-charging the bank. The SELF command is then issued after inserting idle cycles of which the number is specified by bits WTRP1 and WTRP0 in CSnWSR. Synchronous DRAM cannot be accessed while self-refreshing. Self-refreshing mode is cleared by clearing the RMODE bit to 0. After self-refreshing mode has been cleared, command issuance is disabled for the number of cycles specified by bits WTRC1 and WTRC0 in CSnWCR. Self-refreshing timing is shown in figure 7.24. Settings must be made immediately after clearing self-refreshing mode so that auto-refreshing is performed at the correct intervals. When self-refreshing is activated from the auto-refreshing mode, only clearing the RMODE bit to 1 resumes auto-refreshing mode. If it takes long time to start the auto-refreshing, setting RTCNT to the value of RTCOR - 1 starts the auto-refreshing immediately. After self-refreshing has been set, the self-refreshing mode continues even in standby mode, and is maintained even after recovery from standby mode by an interrupt. Since the BSC registers are initialized at a power-on reset, the self-refreshing mode is cleared.
Rev. 6.00 Jun. 12, 2007 Page 176 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Tp CKIO CKE A25 to A0 A11* CSn RAS CAS RD/WR DQMxx D15 to D0 BS
Tpw
Trr
Trc
Trc
Trc
Trc
Trc
Hi-z
Note: * Address pin to be connected to pin A10 of SDRAM.
Figure 7.24 Self-Refreshing Timing Relationship between Refresh Requests and Bus Cycles: If a refresh request occurs during bus cycle execution, the refresh cycle must wait for the bus cycle to be completed. If a new refresh request occurs while the previous refresh request is not performed, the previous refresh request is deleted. To refresh correctly, a bus cycle longer than the refresh interval or the bus busy must be prevented. Power-On Sequence: In order to use synchronous DRAM, mode setting must first be performed after turning the power on. To perform synchronous DRAM initialization correctly, the BSC registers must first be set, followed by writing to the synchronous DRAM mode register. When writing to the synchronous DRAM mode register, the address signal value at that time is latched by a combination of the CSn, RAS, CAS, and RD/WR signals. If the value to be set is X, write to the address of X + (H'F8FD5000) in words. In this operation, the data is ignored. To set burst read/single write, burst read/burst write, CAS latency 2 to 3, wrap type = sequential, and burst length 1 supported by the LSI, arbitrary data is written to the addresses shown in table 7.14 in bytes. In this case, 0s are output at the external address pins of A12 or later.
Rev. 6.00 Jun. 12, 2007 Page 177 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Table 7.17 Access Address for SDRAM Mode Register Write * Burst read/single write (burst length 1)
Data Bus Width 16 bits CAS Latency 2 3 Access Address H'F8FD5440 H'F8FD5460 External Address Pin H'0000440 H'0000460
* Burst read/burst write (burst length 1)
Data Bus Width 16 bits CAS Latency 2 3 Access Address H'F8FD5040 H'F8FD5060 External Address Pin H'0000040 H'0000060
Mode register setting timing is shown in figure 7.25. The PALL command (all bank precharge command) is issued first. The REF command (auto-refreshing command) is then issued eight times. The MRS command (mode register write command) is finally issued. Idle cycles, of which number is specified by bits WTRP1 and WTRP0 in CSnWCR, are inserted between the PALL and the first REF commands. Idle cycles, of which number is specified by bits WTRC1 and WTRC0 in CSnWCR, are inserted between the REF and REF commands, and between the 8th REF and MRS commands. In addition, one or more idle cycles are inserted between the MRS and the next command. It is necessary to keep idle time of certain cycles for SDRAM before issuing the PALL command after turning the power on. Refer the manual of the SDRAM for the idle time to be needed. When the pulse width of the reset signal is longer then the idle time, mode register setting can be started immediately after the reset, but care should be taken when the pulse width of the reset signal is shorter than the idle time.
Rev. 6.00 Jun. 12, 2007 Page 178 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Tp PALL CKIO
Tpw
Trr REF
Trc
Trc
Trr REF
Trc
Trc
Tmw MRS
Tnop
A25 to A0 A11* CSn RAS CAS RD/WR DQMxx D15 to D0 BS Note: * Address pin to be connected to pin A10 of SDRAM. Hi-Z
Figure 7.25 Write Timing for SDRAM Mode Register (Based on JEDEC) 7.5.6 Byte-Selection SRAM Interface
The byte-selection SRAM interface is for access to SRAM which has a byte-selection pin (WEn (BEn)). This interface is used to access to SRAM which has 16-bit data pins and upper and lower byte selection pins, such as UB and LB. When the BAS bit in CSnWCR is cleared to 0 (initial value), the write access timing of the byte-selection SRAM interface is the same as that for the normal space interface. While in read access of a byte-selection SRAM interface, the byte-selection signal is output from the WEn (BEn) pin, which is different from that for the normal space interface. The basic access timing is shown in figure 7.26. In write access, data is written to the memory according to the timing of the byte-selection pin (WEn (BEn)). For details, refer to the data sheet for the corresponding memory. If the BAS bit in CSnWCR is set to 1, the WEn (BEn) pin and RD/WR pin timings change. The basic access timing is shown in figure 7.27. In write access, data is written to the memory according to the timing of the write enable pin (RD/WR). The data hold timing from RD/WR negation to data write must be secured by setting bits HW1 to HW0 in CSnWCR. Figure 7.28 shows the access timing when a software wait is specified.
Rev. 6.00 Jun. 12, 2007 Page 179 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
T1
T2
CKIO
A25 to A0
CSn WEn(BEn)
RD/WR
Read
RD
D15 to D0
RD/WR High
Write
RD
D15 to D0
BS
Figure 7.26 Basic Access Timing for Byte-Selection SRAM (BAS = 0)
Rev. 6.00 Jun. 12, 2007 Page 180 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
T1 CKIO
T2
A25 to A0
CSn WEn(BEn)
RD/WR
Read
RD
D15 to D0
RD/WR Write High RD D15 to D0
BS
Figure 7.27 Basic Access Timing for Byte-Selection SRAM (BAS = 1)
Rev. 6.00 Jun. 12, 2007 Page 181 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Th
T1
Tw
T2
Th
CKIO
A25 to A0
CSn
WEn(BEn)
RD/WR
Read
RD D31 to D0
RD/WR High
Write
RD
D31 to D0
BS
Figure 7.28 Wait Timing for Byte-Selection SRAM (BAS = 1) (Software Wait Only)
Rev. 6.00 Jun. 12, 2007 Page 182 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
This LSI
64 kwords x 16 bits SRAM
...
A1 CSn RD RD/WR D15
...
D0 WE1(BE1) WE0(BE0)
Figure 7.29 Example of Connection with 16-Bit Data-Width Byte-Selection SRAM 7.5.7 PCMCIA Interface
With this LSI, if address map 2 is selected using the MAP bit in CMNCR, the PCMCIA interface can be specified in areas 5 and 6. Areas 5 and 6 in the physical space can be used for the IC memory card and I/O card interface defined in the JEIDA specifications version 4.2 (PCMCIA2.1 Rev. 2.1) by specifying bits TYPE3 to TYPE0 in CSnBCR (n = 5B and 6B) to B'0101. In addition, bits SA1 and SA0 in CSnWCR (n = 5B and 6B) assign the upper or lower 32 Mbytes of each area to an IC memory card or I/O card interface. For example, if bits SA1 and SA0 in CS5BWCR are set to 1 and cleared to 0, respectively, the upper 32 Mbytes and the lower 32 Mbytes of area 5B are used as an IC memory card interface and I/O card interface, respectively. When the PCMCIA interface is used, the bus size must be specified as 8 bits or 16 bits using bits BSZ1 and BSZ0 in CS5BBCR or CS6BBCR. Figure 7.30 shows an example of a connection between this LSI and the PCMCIA card. To enable insertion and removal of the PCMCIA card with the system power turned on, tri-state buffers must be connected between the LSI and the PCMCIA card. In the JEIDA and PCMCIA standards, operation in big endian mode is not clearly defined. Consequently, the provided PCMCIA interface in big endian mode is available only for this LSI.
Rev. 6.00 Jun. 12, 2007 Page 183 of 610 REJ09B0131-0600
...
...
A16
A16 A1 CS OE WE I/O 15 I/O 0 UB LB
Section 7 Bus State Controller (BSC)
This LSI A25 to A0 D7 to D0 D15 to D8 RD/WR CE1A CE2A
G DIR G
PC card (memory or I/O) A25 to A0
D7 to D0
D15 to D8
G DIR
CE1 CE2 RD WE ICIORD ICIOWR I/O Port
G
OE WE/PGM IORD IOWR REG
WAIT IOIS16
Card detection circuit
WAIT IOIS16 CD1, CD2
Figure 7.30 Example of PCMCIA Interface Connection
Rev. 6.00 Jun. 12, 2007 Page 184 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Basic Timing for Memory Card Interface: Figure 7.31 shows the basic timing of the PCMCIA IC memory card interface. If areas 5 and 6 in the physical space are specified as the PCMCIA interface, accessing the common memory areas in areas 5 and 6 automatically accesses with the IC memory card interface. If the external bus frequency (CKIO) increases, the setup times and hold times for the address pins (A25 to A0), card enable signals (CE1A, CE2A, CE1B, CE2B), and write data (D15 to D0) to the RD and WE signals become insufficient. To prevent this error, this LSI can specify the setup times and hold times for areas 5 and 6 in the physical space independently, using CS5BWCR and CS6BWCR. In the PCMCIA interface, as in the normal space interface, a software wait or hardware wait can be inserted using the WAIT pin. Figure 7.32 shows the PCMCIA memory bus wait timing.
Tpcm1 CKIO
Tpcm1w
Tpcm1w
Tpcm1w
Tpcm2
A25 to A0 CExx RD/WR RD Read D15 to D0 WE Write D15 to D0 BS
Figure 7.31 Basic Access Timing for PCMCIA Memory Card Interface
Rev. 6.00 Jun. 12, 2007 Page 185 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Tpcm0 CKIO A25 to A0 CExx RD/WR RD Read D15 to D0 WE Write D15 to D0 BS WAIT
Tpcm0w
Tpcm1
Tpcm1w
Tpcm1w Tpcm1w
Tpcm1w
Tpcm2
Tpcm2w
Figure 7.32 Wait Timing for PCMCIA Memory Card Interface (TED[3:0] = B'0010, TEH[3:0] = B'0001, Software Wait = 1, Hardware Wait = 1) When 32 Mbytes of the memory space are used as an IC memory card interface, a port is used to generate the REG signal that switches between the common memory and attribute memory. When the memory space used for the IC memory card interface is 16 Mbytes or less, pin A24 can be used as the REG signal by allocating a 16-Mbyte common memory space and a 16-Mbyte attribute memory space alternatively.
Rev. 6.00 Jun. 12, 2007 Page 186 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
PCMCIA interface area is 32 Mbytes (An I/O port pin is used as the REG) Area 5 : H'14000000 Attribute memory/common memory Area 5 : H'16000000 I/O space Area 6 : H'18000000 Attribute memory/common memory Area 6 : H'1A000000 I/O space
PCMCIA interface area is 16 Mbytes (A24 is used as the REG) Area 5 : H'14000000 Area 5 : H'15000000 Area 5 : H'16000000 H'17000000 Area 6 : H'18000000 Area 6 : H'19000000 Area 6 : H'1A000000 H'1B000000 Attribute memory Common memory I/O space Attribute memory Common memory I/O space
Figure 7.33 Example of PCMCIA Space Assignment (CS5BWCR.SA[1:0] = B'10, CS6BWCR.SA[1:0] = B'10) Basic Timing for I/O Card Interface: Figures 7.34 and 7.35 show the basic timings for the PCMCIA I/O card interface. The I/O card and IC memory card interfaces are specified by an address to be accessed. When area 5 of the physical space is specified as the PCMCIA and both bits SA1 and SA0 in CS5BWCR are set to 1, the I/O card interface can automatically be specified by accessing the physical addresses from H'16000000 to H'17FFFFFF and from H'14000000 to H'15FFFFFF. When area 6 of the physical space is specified as the PCMCIA and both bits SA1 and SA0 in CS6BWCR are set to 1, the I/O card interface can automatically be specified by accessing the physical addresses from H'1A000000 to H'1BFFFFFF and from H'18000000 to H'19FFFFFF. Note that areas to be accessed as the PCMCIA I/O card must be non-cached (space P2). If the PCMCIA card is accessed as an I/O card in little endian mode, dynamic bus sizing for the I/O bus can be achieved using the IOIS16 signal. If the IOIS16 signal is driven high in a word-size I/O bus cycle while the bus width of area 6 is specified as 16 bits, the bus width is recognized as 8 bits and data is accessed twice in units of eight bits in the I/O bus cycle to be executed.
Rev. 6.00 Jun. 12, 2007 Page 187 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
The IOIS16 signal is sampled at the falling edge of the CKIO signal in the Tpci0, Tpci0w, and Tpci1 cycles when bits TED3 to TED0 are specified as 1.5 cycles or more, and is reflected in the CE2 signal 1.5 cycles after the CKIO sampling point. Bits TED3 to TED0 must be specified appropriately to satisfy the setup time of the PC card from ICIORD and ICIOWR to CEn. Figure 7.36 shows the dynamic bus sizing basic timing. Note that the IOIS16 signal is not supported in big endian mode. In the big endian mode, the IOIS16 signal must be fixed low.
Tpci1 Tpci1w Tpci1w Tpci1w Tpci2
CKIO
A25 to A0
CExx
RD/WR
ICIORD Read D15 to D0
ICIOWR Write D15 to D0
BS
Figure 7.34 Basic Timing for PCMCIA I/O Card Interface
Rev. 6.00 Jun. 12, 2007 Page 188 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Tpci0 CKIO A25 to A0 CExx RD/WR ICIORD Read D15 to D0 ICIOWR Write D15 to D0 BS WAIT IOIS16
Tpci0w
Tpci1
Tpci1w
Tpci1w
Tpci1w
Tpci1w
Tpci2
Tpci2w
Figure 7.35 Wait Timing for PCMCIA I/O Card Interface (TED[3:0] = B'0010, TEH[3:0] = B'0001, Software Wait = 1, Hardware Wait = 1)
Tpci0 CKIO A25 to A0 CE1x CE2x RD/WR ICIORD Read D15 to D0 ICIOWR Write D15 to D0 BS WAIT IOIS16
Tpci0w
Tpci1
Tpci1w
Tpci1w
Tpci1w
Tpci1w
Tpci2
Tpci2w
Tpci0
Tpci0w
Tpci1
Tpci1w
Tpci1w
Tpci1w
Tpci1w
Tpci2
Tpci2w
Figure 7.36 Timing for Dynamic Bus Sizing of PCMCIA I/O Card Interface (TED[3:0] = B'0010, TEH[3:0] = B'0001, Software Waits = 3)
Rev. 6.00 Jun. 12, 2007 Page 189 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
7.5.8
Wait between Access Cycles
Data output in the previous cycle may conflict with that in the next cycle because the buffer-off timing of devices with slow access speed cannot be operated to satisfy the higher operating frequency of LSIs. As a result of these conflict, the reliability of the device is low and malfunctions may occur. This LSI has a function that avoids data conflicts by inserting wait cycles between consecutive access cycles. The number of wait cycles between access cycles can be set by bits IWW1 and IWW0, bits IWRWD1 and IWRWD0, bits IWRWS1 and IWRWS0, bits IWRRD1 and IWRRD0, and bits IWRRS1 and IWRRS0 in CSnBCR. The conditions for setting the wait cycles between access cycles (idle cycles) are shown below. 1. 2. 3. 4. 5. Consecutive accesses are write-read or write-write Consecutive accesses are read-write for different areas Consecutive accesses are read-write for the same area Consecutive accesses are read-read for different areas Consecutive accesses are read-read for the same area Others
7.5.9
Reset: The bus state controller (BSC) can be initialized completely only by a power-on reset. At power-on reset, all signals are negated and output buffers are turned off regardless of the bus cycle state. All control registers are initialized. In standby mode and sleep mode, control registers of the BSC are not initialized. Some flash memories may stipulate a minimum time from reset release to the first access. To ensure this minimum time, the BSC supports a 7-bit counter (RWTCNT). At a power-on reset, the RWTCNT contents are cleared to 0. After a power-on reset, RWTCNT is counted up in synchronization with the CKIO signal and an external access will not be generated until RWTCNT is counted up to H'007F. Access from the Site of the LSI Internal Bus Master: There are three types of LSI internal buses: a cache bus, internal bus, and peripheral bus. The CPU and cache memory are connected to the cache bus. Internal bus masters other than the CPU and BSC are connected to the internal bus. Low-speed peripheral modules are connected to the peripheral bus. Internal memory other than the cache memory and debugging modules such as the UBC are connected to both the cache bus and internal bus. Access from the cache bus to the internal bus is enabled but access from the internal bus to the cache bus is disabled. This gives rise to the following problems.
Rev. 6.00 Jun. 12, 2007 Page 190 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
Internal bus masters other than the CPU such as the E-DMAC can access on-chip memory other than the cache memory but cannot access the cache memory. If an on-chip bus master other than the CPU writes data to an external memory other than the cache, the contents of the external memory may differ from that of the cache memory. To prevent this problem, if the external memory whose contents is cached is written by an on-chip bus master other than the CPU, the corresponding cache memory should be purged by software. If the CPU initiates read access for the cache, the cache is searched. If the cache stores data, the CPU latches the data and completes the read access. If the cache does not store data, the CPU performs four consecutive longword read cycles to perform cache fill operations via the internal bus. If a cache miss occurs in byte or word operand access or at a branch to an odd word boundary (4n + 2), the CPU performs four consecutive longword accesses to perform a cache fill operation on the external interface. For a cache-through area, the CPU performs access according to the actual access addresses. For an instruction fetch to an even word boundary (4n), the CPU performs longword access. For an instruction fetch to an odd word boundary (4n + 2), the CPU performs word access. For a read cycle of a cache-through area or an on-chip peripheral module, the read cycle is first accepted and then read cycle is initiated. The read data is sent to the CPU via the cache bus. In a write cycle for the cache area, the write cycle operation differs according to the cache write methods. In write-back mode, the cache is first searched. If data is detected at the address corresponding to the cache, the data is then re-written to the cache. In the actual memory, data will not be re-written until data in the corresponding address is re-written. If data is not detected at the address corresponding to the cache, the cache is updated. In this case, data to be updated is first saved to the internal buffer, 16-byte data including the data corresponding to the address is then read, and data in the corresponding access of the cache is finally updated. Following these operations, a write-back cycle for the saved 16-byte data is executed. In write-through mode, the cache is first searched. If data is detected at the address corresponding to the cache, the data is re-written to the cache simultaneously with the actual write via the internal bus. If data is not detected at the address corresponding to the cache, the cache is not updated but an actual write is performed via the internal bus. Since the BSC incorporates a 1-stage write buffer, the BSC can execute an access via the internal bus before the previous external bus cycle is completed in a write cycle. If the on-chip module is read or written after the external low-speed memory is written, the on-chip module can be accessed before the completion of the external low-speed memory write cycle.
Rev. 6.00 Jun. 12, 2007 Page 191 of 610 REJ09B0131-0600
Section 7 Bus State Controller (BSC)
In read cycles, the CPU is placed in the wait cycle until read operation has been completed. To continue the process after the data write to the device has been completed, perform a dummy read to the same address to check for completion of the write before the next process to be executed. The write buffer of the BSC functions in the same way for an access by a bus master other than the CPU such as the DMAC or E-DMAC. Accordingly, to perform dual address DMA transfers, the next read cycle is initiated before the previous write cycle is completed. Note, however, that if both the DMA source and destination addresses exist in external memory space, the next write cycle will not be initiated until the previous write cycle is completed. On-Chip Peripheral Module Access: To access an on-chip module register, two or more peripheral module clock (P) cycles are required. Care must be taken in system design.
Rev. 6.00 Jun. 12, 2007 Page 192 of 610 REJ09B0131-0600
Section 8 Clock Pulse Generator (CPG)
Section 8 Clock Pulse Generator (CPG)
This LSI has a clock pulse generator (CPG) that generates an internal clock (I), a peripheral clock (P), a bus clock (B) and a clock (M) for an IEEE802.3-PHY (physical layer device), hereinafter called PHY-LSI. The CPG consists of an oscillator, PLL circuits, and divider circuits.
8.1
Features
* Four clock modes Selection of four clock modes depending on the frequency of a clock source and whether a crystal resonator or external clock input is in use. * Four clocks generated independently An internal clock (I) for the CPU and cache; a peripheral clock (P) for the on-chip peripheral modules; a bus clock (B = CKIO) for the external bus interface; and a clock (M) for the PHY-LSI. * Frequency change function Frequencies of the internal clock, peripheral clock, and clock for the PHY-LSI can be changed independently using the PLL circuit and divider circuit within the CPG. Frequencies are changed by software using the frequency control register (FRQCR) and PHY-LSI clock frequency control register (MCLKCR) settings. * Power-down mode control The clock can be stopped in sleep mode and software standby mode and specific modules can be stopped using the module standby function.
CPGS301C_000020030900
Rev. 6.00 Jun. 12, 2007 Page 193 of 610 REJ09B0131-0600
Section 8 Clock Pulse Generator (CPG)
A block diagram of the CPG is shown in figure 8.1.
Oscillator unit
CK-PHY Divider 2 x1 x1/2 x1/4 Divider 1 x1 PLL circuit 1 (x1, x2) CKIO XTAL EXTAL Crystal oscillator Bus clock (B = CKIO) x1/2 x1/4 PHY clock (M)
Internal clock (I)
PLL circuit 2 (x2, x4)
Peripheral clock (P)
CPG control unit MD2 MD1 MD0 FRQCR Clock frequency control circuit
Standby control circuit
MCLKCR
STBCR
STBCR2
STBCR3
STBCR4
Bus interface
[Legend] FRQCR: STBCR: STBCR2: STBCR3: STBCR4: MCLKCR:
Internal bus Frequency control register Standby control register Standby control register 2 Standby control register 3 Standby control register 4 PHY-LSI clock frequency control register
Figure 8.1 Block Diagram of CPG
Rev. 6.00 Jun. 12, 2007 Page 194 of 610 REJ09B0131-0600
Section 8 Clock Pulse Generator (CPG)
The clock pulse generator blocks function as follows: PLL Circuit 1: PLL circuit 1 leaves the input clock frequency from the PLL circuit 2 unchanged or doubles it. The multiplication ratio is set by the frequency control register. The phase of the rising edge of the internal clock is controlled so that it will match the phase of the rising edge of the CKIO pin. PLL Circuit 2: PLL circuit 2 doubles or quadruples the clock frequency input from the crystal oscillator or the EXTAL pin. The multiplication ratio is fixed for each clock operating mode. The clock operating mode is set with pins MD0, MD1, or MD2. Crystal Oscillator: The crystal oscillator is an oscillator circuit when a crystal resonator is connected to the XTAL and EXTAL pins. The crystal oscillator can be used by setting the clock operating mode. Divider 1: Divider 1 generates clocks with the frequencies used by the internal clock, peripheral clock, and bus clock. The frequency output as the internal clock is always the same as that of the devider1 output. The frequency output as the bus clock is automatically selected so that it is the same as the frequency of the CKIO signal according to the multiplication ratio of PLL circuit 1. The frequencies can be 1, 1/2, or 1/4 times the output frequency of PLL circuit 1, as long as it stays at or above the frequency of the CKIO pin. The division ratio is set in the frequency control register. Divider 2: Divider 2 generates a clock that is supplied to the external PHY-LSI. Divider 2 must output 25-MHz frequency for the PHY-LSI that generally requires 25-MHz clock. The output clock of divider 2 can be 1, 1/2, or 1/4 times the output frequency of PLL circuit 1. The division ratio is set in the PHY-LSI clock frequency control register. Clock Frequency Control Circuit: The clock frequency control circuit controls the clock frequency using pins MD0, MD1, and MD2, the frequency control register, and PHY-LSI clock frequency control register. Standby Control Circuit: The standby control circuit controls the state of the on-chip oscillator circuit and other modules during clock switching and in software standby mode. Frequency Control Register: The frequency control register has control bits assigned for the following functions: clock output/non-output from the CKIO pin, the frequency multiplication ratio of PLL circuit 1, and the frequency division ratio of the peripheral clock. Standby Control Register: The standby control register has bits for controlling the power-down modes. For details, see section 10, Power-Down Modes.
Rev. 6.00 Jun. 12, 2007 Page 195 of 610 REJ09B0131-0600
Section 8 Clock Pulse Generator (CPG)
PHY-LSI Clock Frequency Control Register: The PHY-LSI clock frequency control register sets the frequency division ratio of the PHY-LSI clock.
8.2
Input/Output Pins
Table 8.1 shows the CPG pin configuration. Table 8.1
Pin Name Mode control pins*
Pin Configuration
Abbr. MD0 MD1 MD2 I/O Input Input Input Output Input Output Description Set the clock operating mode. Set the clock operating mode. Set the clock operating mode. Connects a crystal resonator. Connects a crystal resonator or an external clock. Outputs an external clock. Outputs a clock for an external PHY-LSI.
Clock input pins
XTAL EXTAL
Clock output pin PHY-LSI clock pin Note: *
CKIO
CK_PHY Output
The values of these mode control pins are sampled only at a power-on reset or in a software standby with the MDCHG bit in STBCR to 1. This can prevent the erroneous operation of this LSI.
8.3
Clock Operating Modes
Table 8.2 shows the relationship between the mode control pins (MD2 to MD0) combinations and the clock operating modes. Table 8.3 shows the usable frequency ranges in the clock operating modes and the frequency range of the input clock. Table 8.2 Mode Control Pins and Clock Operating Modes
Clock I/O Source EXTAL Crystal resonator EXTAL Crystal resonator Output CKIO CKIO CKIO CKIO PLL2 ON (x4) ON (x4) ON (x2) ON (x2) PLL1 ON (x1, x2) ON (x1, x2) ON (x1, x2) ON (x1, x2) CKIO Frequency (EXTAL) x 4 (Crystal resonator) x 4 (EXTAL) x 2 (Crystal resonator) x 2
Clock Pin Values Operating Mode MD2 MD1 MD0 1 2 5 6 0 0 1 1 0 1 0 1 1 0 1 0
Rev. 6.00 Jun. 12, 2007 Page 196 of 610 REJ09B0131-0600
Section 8 Clock Pulse Generator (CPG)
Mode 1: The frequency of the external clock input from the EXTAL pin is quadrupled by PLL circuit 2, and then the clock is supplied to this LSI. Since the input clock frequency ranging 10 MHz to 12.5 MHz can be used, the CKIO frequency ranges from 40 MHz to 50 MHz. Mode 2: The frequency of the on-chip crystal oscillator output is quadrupled by PLL circuit 2, and then the clock is supplied to this LSI. Since the crystal resonator frequency ranging 10 MHz to 12.5 MHz can be used, the CKIO frequency ranges from 40 MHz to 50 MHz. Mode 5: The frequency of the external clock from the EXTAL pin is doubled by PLL circuit 2, and then the clock is supplied to this LSI. Since the input clock frequency ranging 10 MHz to 25 MHz, the CKIO frequency ranges from 20 MHz to 50 MHz. Mode 6: The frequency of the on-chip crystal oscillator output is doubled by PLL circuit 2, and then the clock is supplied to the LSI. Since the crystal oscillation frequency ranging 10 MHz to 25 MHz can be used, the CKIO frequency ranges from 20 MHz to 50 MHz. Table 8.3 Possible Combination of Clock Modes and FRQCR Values
FRQCR Register Value H'1000 H'1001 H'1003 H'1101 H'1103 5 or 6 H'1000 H'1001 H'1003 H'1101 H'1103 Note: * PLL Circuit 1 ON (x1) ON (x1) ON (x1) ON (x2) ON (x2) ON (x1) ON (x1) ON (x1) ON (x2) ON (x2) PLL Circuit 2 ON (x4) ON (x4) ON (x4) ON (x4) ON (x4) ON (x2) ON (x2) ON (x2) ON (x2) ON (x2) Clock Ratio* (I:B:P) 4:4:4 4:4:2 4:4:1 8:4:4 8:4:2 2:2:2 2:2:1 2:2:1/2 4:2:2 4:2:1 10 MHz to 25 MHz 20 MHz to 50 MHz Input Clock Frequency Range 10 MHz to 12.5 MHz CKIO Pin Frequency Range 40 MHz to 50 MHz
Mode 1 or 2
Input clock is assumed to be 1.
Rev. 6.00 Jun. 12, 2007 Page 197 of 610 REJ09B0131-0600
Section 8 Clock Pulse Generator (CPG)
Cautions: 1. The internal clock frequency is the product of the frequency of the CKIO pin and the frequency multiplication ratio of PLL circuit 1. 2. The peripheral clock frequency is the product of the frequency of the CKIO pin, the frequency multiplication ratio of PLL circuit 1, and the division ratio of divider 1. Do not set the peripheral clock frequency lower than the CKIO pin frequency. 3. The PHY-LSI clock frequency is the product of the frequency of the CKIO pin, the frequency multiplication ratio of PLL circuit 1, and the division ratio of divider 2. 4. x1, x1/2, or x1/4 can be used as the division ratio of divider 1. Set the rate in the frequency control register. 5. The division ratio of divider 2 is selected from x1, x1/2, or x1/4 can be used as Set the rate in the PHY-LSI clock frequency control register. 6. The output frequency of PLL circuit 1 is the product of the frequency of the CKIO pin and the multiplication ratio of PLL circuit 1. It is set by the frequency control register. 7. The bus clock frequency is always set to be equal to the frequency of the CKIO pin. 8. The clock mode, the FRQCR register value, and the frequency of the input clock should be decided to satisfy the range of operating frequency specified in section 21, Electrical Characteristics, with referring to table 8.3.
8.4
Register Descriptions
The CPG has the following registers. For details on the addresses of these registers and the states of these registers in each processing state, see section 20, List of Registers. * Frequency control register (FRQCR) * PHY-LSI clock frequency control register (MCLKCR) 8.4.1 Frequency Control Register (FRQCR)
FRQCR is a 16-bit readable/writable register that specifies whether a clock is output from the CKIO pin in standby mode, the frequency multiplication ratio of PLL circuit 1, and the frequency division ratio of the peripheral clock. Only word access can be used on FRQCR. FRQCR is initialized by a power-on reset due to the external input signal. However, it is not initialized by a power-on reset due to a WDT overflow.
Rev. 6.00 Jun. 12, 2007 Page 198 of 610 REJ09B0131-0600
Section 8 Clock Pulse Generator (CPG)
Bit 15 to 13
Bit Name
Initial Value All 0
R/W R
Description Reserved These bits are always read as 0. The write value should always be 0.
12
CKOEN
1
R/W
Clock Output Enable Specifies whether a clock continues to be output from the CKIO pin or the output level of the CKIO signal is fixed when leaving software standby mode. The CKIO output is fixed low when this bit is set to 0. Therefore, the malfunction of external circuits because of an unstable CKIO clock when leaving software standby mode can be prevented. 0: Output level of the CKIO signal is fixed low in software standby mode. 1: Clock input to the EXTAL pin is output to the CKIO pin during software standby mode in clock mode 1 or 5. However, the output level of the CKIO pin is fixed low for two cycles of P when changing from the normal mode to the standby mode. This prevents hazard which occurs when the source of the CKIO signal is changed from the PLL2 output to the EXTAL signal.
11
0
R
Reserved This bit is always read as 0. The write value should always be 0.
10 9 8 7 to 3
STC2 STC1 STC0
0 0 0 All 0
R/W R/W R/W R
PLL Circuit 1 Frequency Multiplication Ratio 000: x1 001: x2 Other values: Setting prohibited Reserved These bits are always read as 0. The write value should always be 0.
Rev. 6.00 Jun. 12, 2007 Page 199 of 610 REJ09B0131-0600
Section 8 Clock Pulse Generator (CPG)
Bit 2 1 0
Bit Name PFC2 PFC1 PFC0
Initial Value 0 1 1
R/W R/W R/W R/W
Description Peripheral Clock Frequency Division Ratio Specify the division ratio of the peripheral clock frequency with respect to the output frequency of PLL circuit 1. 000: x1 001: x1/2 011: x1/4 Other values: Setting prohibited
8.4.2
PHY-LSI Clock Frequency Control Register (MCLKCR)
MCLKCR is an 8-bit readable/writable register. This register must be written to in words. The upper byte of the word data must be H'5A and the lower byte is the write data.
Bit 7 6 Bit Name FLSCS1 FLSCS0 Initial Value 0 1 R/W R/W R/W Description Source Clock Select Select the source clock. 00: PLL1 output clock 01: PLL1 output clock 10: Setting prohibited 11: Setting prohibited 5 to 3 All 0 R Reserved These bits are always read as 0. The write value should always be 0. 2 1 0 FLDIVS2 FLDIVS1 FLDIVS0 0 1 1 R/W R/W R/W Divider Select Set the division ratio of PLL1 output. 000: x1 001: x1/2 011: x1/4 Other values: Setting prohibited
Rev. 6.00 Jun. 12, 2007 Page 200 of 610 REJ09B0131-0600
Section 8 Clock Pulse Generator (CPG)
8.4.3
Usage Notes
* MCLKCR is used only for generation of external PHY-LSI clocks. * When changing the contents of MCLKCR or FRQCR, make sure that the external PHY-LSI is in the reset state. Otherwise, a hazard may be generated on the CK_PHY signal. After the contents of MCLKCR or FRQCR have been changed, clear the reset state of the PHY-LSI.
Rev. 6.00 Jun. 12, 2007 Page 201 of 610 REJ09B0131-0600
Section 8 Clock Pulse Generator (CPG)
8.5
Changing Frequency
The internal clock frequency can be changed by changing the multiplication ratio of PLL circuit 1. The peripheral clock frequency can be changed either by changing the multiplication ratio of PLL circuit 1 or by changing the division ratio of divider 1. All of these are controlled by software through the frequency control register. The methods are described below. 8.5.1 Changing Multiplication Ratio
The PLL lock time must be preserved when the multiplication ratio of PLL circuit 1 is changed. The on-chip WDT counts for preserving the PLL lock time. 1. In the initial state, the multiplication ratio of PLL circuit 1 is 1. 2. Set a value that satisfies the given PLL lock time in the WDT and stop the WDT. The following must be set. TME bit in WTCSR = 0: WDT stops Bits CKS2 to CKS0 in WTCSR: Division ratio of WDT count clock WTCNT: Initial counter value 3. Set the desired value in bits STC2 to STC0 while the MDCHG bit in STBCR is 0. The division ratio can also be set in bits PFC2 to PFC0. 4. This LSI pauses internally and the WDT starts incrementing. The internal and peripheral clocks both stop and only the WDT is supplied with the clock. The clock will continue to be output on the CKIO pin. 5. Supply of the specified clock starts at a WDT count overflow, and this LSI starts operating again. The WDT stops after it overflows. Notes: 1. When the MDCHG bit in STBCR is set to 1, changing the FRQCR value has no effect on the operation immediately. For details, see section 8.5.3, Changing Clock Operating Mode. 2. The multiplication ratio should be changed after completion of the operation, if the onchip peripheral module is operating. The internal and peripheral clocks are stopped during the multiplication ratio is changed. The communication error may occur by the peripheral module communicating to the external IC, and the time error may occur by the timer unit (except the WDT). The edge detection of external interrupts (NMI and IRQ7 to IRQ0) cannot be performed.
Rev. 6.00 Jun. 12, 2007 Page 202 of 610 REJ09B0131-0600
Section 8 Clock Pulse Generator (CPG)
8.5.2
Changing Division Ratio
The WDT will not count unless the multiplication ratio is changed simultaneously. 1. In the initial state, PFC2 to PFC0 = 011. 2. Set the desired values in bits PFC2 to PFC0 while the MDCHG bit in STBCR is 0. The values that can be set are limited by the clock mode and the multiplication ratio of PLL circuit 1. Note that if the wrong value is set, this LSI will malfunction. 3. The clock is immediately changed to the new division ratio. Note: When the MDCHG bit in STBCR is set to 1, changing the FRQCR value has no effect on the operation immediately. For details, see section 8.5.3, Changing Clock Operating Mode. 8.5.3 Changing Clock Operating Mode
The values of the mode control pins (MD2 to MD0) that define a clock operating mode are fetched at a power-on reset and software standby while the MDCHG bit in STBCR is set to 1 register. Even if changing the FRQCR with the MDCHG bit set to 1, the clock mode cannot immediately be changed to the specified clock mode. This change can be reflected as a multiplication ratio or a division ratio after leaving software standby mode to change operating modes. Reducing the PLL settling time without changing again the multiplication ratio after the operating mode changing is possible by the use of this. The procedures for the mode change using software standby mode are described below. 1. Set bits MD2 to MD0 to the desired clock operating mode. 2. Set both the STBY and MDCHG bits in STBCR to 1. 3. Set the adequate value to the WDT so that the given oscillation settling time can be satisfied. Then stop the WDT. 4. Set FRQCR to the desired mode. Set bits STC2 to STC0 to the desired multiplication ratio. At this time, a division ratio can be set in bits PFC2 to PFC0. During the operation before the mode change, the clock cannot be changed to the specified clock. 5. Enter software standby mode using the SLEEP instruction. 6. Leave software standby mode using an interrupt. 7. After leaving software standby mode, this LSI starts the operation with the value of FRQCR that has been set before the mode change.
Rev. 6.00 Jun. 12, 2007 Page 203 of 610 REJ09B0131-0600
Section 8 Clock Pulse Generator (CPG)
Notes: 1. Pins MD2 to MD0 should be set during the operation before the mode change or during software standby mode before requesting an interrupt. 2. Clear the STBY bit in STBCR in the exception handling routine for the interrupt in step 6. Otherwise, software standby mode is entered again. For details, see section 10.5.2, Canceling Software Standby Mode. 3. Once bits STC2 to STC0 are changed, the clock is not switched to the specified clock even if only bits PFC2 to PFC0 are changed. When bits STC2 to STC0 are changed after the MDCHG bit has been set to 1, the FRQCR setting must not be made until the clock mode is changed.
Rev. 6.00 Jun. 12, 2007 Page 204 of 610 REJ09B0131-0600
Section 8 Clock Pulse Generator (CPG)
8.6
Notes on Board Design
When Using an External Crystal Resonator: Place the crystal resonator, capacitors CL1 and CL2, and damping resistor R close to the EXTAL and XTAL pins. To prevent induction from interfering with correct oscillation, use a common grounding point for the capacitors connected to the resonator, and do not locate a wiring pattern near these components.
Avoid crossing signal lines
CL1
CL2
R EXTAL XTAL Note : Although the recommended value for CL1 and CL2 is 20pF and the recommended value for R is 0, the values should be determined after consultation with the crystal manufacturer.
This LSI
Figure 8.2 Points for Attention when Using Crystal Resonator Bypass Capacitors: Insert a laminated ceramic capacitor as a bypass capacitor for each VSS/VSSQ and VCC/VCCQ pair. Mount the bypass capacitors to the power supply pins, and use components with a frequency characteristic suitable for the operating frequency of the LSI, as well as a suitable capacitance value. * Digital power supply pairs for internal logic A4-B4, B11-A11, D15-D14, E2-E1, G12-G13, H4-H3, J12-J13, M1-M2, M8-N8, P5-R5 * Power supply pairs for input and output A1-B1, A7-B7, A15-A14, F15-F14, K1-K2, M12-P14, M15-M14, R1-R2, R10-P10 * Power supply pairs for PLL N13-N14, N13-P15
Rev. 6.00 Jun. 12, 2007 Page 205 of 610 REJ09B0131-0600
Section 8 Clock Pulse Generator (CPG)
When Using a PLL Oscillator Circuit: Keep the wiring from the PLL VCC and PLL VSS connection pattern to the power supply pins short, and make the pattern width large, to minimize the inductance component. The analog power supply system of the PLL is sensitive to a noise. Therefore, the system malfunction may occur by the intervention with other power supply. Do not supply the analog power supply with the same resource as the digital power supply of VCC and VCCQ.
Rev. 6.00 Jun. 12, 2007 Page 206 of 610 REJ09B0131-0600
Section 9 Watchdog Timer (WDT)
Section 9 Watchdog Timer (WDT)
This LSI includes the watchdog timer (WDT) that can reset this LSI by the overflow of the counter when the value of the counter has not been updated because of a system runaway. The WDT is a single-channel timer that uses a peripheral clock as an input and counts the clock settling time when leaving software standby mode and temporary standby state, such as frequency changes. It can also be used as an interval timer.
9.1
Features
The WDT has the following features: * Can be used to ensure the clock settling time. The WDT can be used when leaving software standby mode and the temporary standby state which occur when the clock frequency is changed. * Can switch between watchdog timer mode and interval timer mode. * Internal resets in watchdog timer mode Internal resets are generated when the counter overflows. * Interrupts are generated in interval timer mode Interval timer interrupts are generated when the counter overflows. * Choice of eight counter input clocks Eight clocks (x1 to x1/4096) that are obtained by dividing the peripheral clock can be chosen.
WDTS300B_000020030900
Rev. 6.00 Jun. 12, 2007 Page 207 of 610 REJ09B0131-0600
Section 9 Watchdog Timer (WDT)
Figure 9.1 is a block diagram of the WDT.
WDT Standby cancellation Standby mode Peripheral clock (P) Internal reset request Reset control Clock selection Clock selector Interrupt request Interrupt control WTCSR Overflow Clock Divider
Standby control
WTCNT
Bus interface
[Legend] WTCSR: WTCNT: Watchdog timer control/status register Watchdog timer counter
Figure 9.1 Block Diagram of WDT
Rev. 6.00 Jun. 12, 2007 Page 208 of 610 REJ09B0131-0600
Section 9 Watchdog Timer (WDT)
9.2
Register Descriptions
The WDT has the following two registers. For details on the addresses of these registers and the states of these registers in each processing state, see section 20, List of Registers. * Watchdog timer counter (WTCNT) * Watchdog timer control/status register (WTCSR) 9.2.1 Watchdog Timer Counter (WTCNT)
WTCNT is an 8-bit readable/writable register that increments on the selected clock. When an overflow occurs, it generates a reset in watchdog timer mode and an interrupt in interval time mode. WTCNT is not initialized by an internal power-on reset due to the WDT overflow. WTCNT is initialized to H'00 by a power-on reset input to the pin and an H-UDI reset. Use a word access to write to WTCNT, with H'5A in the upper byte. Use a byte access to read WTCNT. Note: The writing method for WTCNT differs from other registers so that the WTCNT value cannot be changed accidentally. For details, see section 9.2.3, Notes on Register Access. 9.2.2 Watchdog Timer Control/Status Register (WTCSR)
WTCSR is an 8-bit readable/writable register composed of bits to select the clock used for the counting, bits to select the timer mode and overflow flags, and enable bits. WTCSR holds its value in the internal reset state due to the WDT overflow. WTCSR is initialized to H'00 by a power-on reset input to the pin and an H-UDI reset. To use it for counting the clock settling time when leaving software standby mode, WTCSR holds its value after a counter overflow. Use a word access to write to WTCSR, with H'A5 in the upper byte. Use a byte access to read WTCSR. Note: The writing method for WTCNT differs from other registers so that the WTCNT value cannot be changed accidentally. For details, see section 9.2.3, Notes on Register Access.
Rev. 6.00 Jun. 12, 2007 Page 209 of 610 REJ09B0131-0600
Section 9 Watchdog Timer (WDT)
Bit 7
Initial Bit Name Value TME 0
R/W R/W
Description Timer Enable Starts and stops timer operation. Clear this bit to 0 when using the WDT in software standby mode or when changing the clock frequency. 0: Timer disabled: Count-up stops and WTCNT value is retained 1: Timer enabled
6
WT/IT
0
R/W
Timer Mode Select Selects whether to use the WDT as a watchdog timer or an interval timer. 0: Interval timer mode 1: Watchdog timer mode Note: If WT/IT is modified when the WDT is operating, the up-count may not be performed correctly.
5
0
R
Reserved This bit is always red as 0. The write value should always be 0.
4
WOVF
0
R/W
Watchdog Timer Overflow Indicates that WTCNT has overflowed in watchdog timer mode. This bit is not set in interval timer mode. 0: No overflow 1: WTCNT has overflowed in watchdog timer mode
3
IOVF
0
R/W
Interval Timer Overflow Indicates that WTCNT has overflowed in interval timer mode. This bit is not set in watchdog timer mode. 0: No overflow 1: WTCNT has overflowed in interval timer mode
Rev. 6.00 Jun. 12, 2007 Page 210 of 610 REJ09B0131-0600
Section 9 Watchdog Timer (WDT)
Bit 2 1 0
Initial Bit Name Value CKS2 CKS1 CKS0 0 0 0
R/W R/W R/W R/W
Description Clock Select 2 to 0 These bits select the clock to be used for the WTCNT count from the eight types obtainable by dividing the peripheral clock (P). The overflow period that is shown inside the parenthesis in the table is the value when the peripheral clock (P) is 25 MHz. 000: P (10 s) 001: P/4 (41 s) 010: P/16 (164 s) 011: P/32 (328 s) 100: P/64 (655 s) 101: P/256 (2.62 ms) 110: P/1024 (10.49 ms) 111: P/4096 (41.94 ms) Note: If bits CKS2 to CKS0 are modified when the WDT is operating, the up-count may not be performed correctly. Ensure that these bits are modified only when the WDT is not operating.
9.2.3
Notes on Register Access
The watchdog timer counter (WTCNT) and watchdog timer control/status register (WTCSR) are more difficult to write to than other registers. The procedure for writing to these registers is given below. Writing to WTCNT and WTCSR: These registers must be written by a word transfer instruction. They cannot be written by a byte or longword transfer instruction. When writing to WTCNT, set the upper byte to H'5A and transfer the lower byte as the write data, as shown in figure 9.2. When writing to WTCSR, set the upper byte to H'A5 and transfer the lower byte as the write data. This transfer procedure writes the lower byte data to WTCNT or WTCSR.
Rev. 6.00 Jun. 12, 2007 Page 211 of 610 REJ09B0131-0600
Section 9 Watchdog Timer (WDT)
WTCNT write
15 8 H'5A 7 Write data 0
Address: H'F815FF84
WTCSR write 15 Address: H'F815FF86 H'A5 8 7 Write data 0
Figure 9.2 Writing to WTCNT and WTCSR
9.3
9.3.1
WDT Operation
Canceling Software Standbys
The WDT can be used to cancel software standby mode with an NMI interrupt or external interrupt (IRQ). The procedure is described below. (The WDT does not run when resets are used for canceling, so keep the RES pin low until the clock stabilizes.) 1. Before transition to software standby mode, always clear the TME bit in WTCSR to 0. When the TME bit is 1, an erroneous reset or interval timer interrupt may be generated when the count overflows. 2. Set the type of count clock used in the CKS2 to CKS0 bits in WTCSR and the initial values for the counter in WTCNT. These values should ensure that the time till count overflow is longer than the clock oscillation settling time. 3. Move to software standby mode by executing a SLEEP instruction to stop the clock. 4. The WDT starts counting by detecting the change of input levels of the NMI or IRQ pin. 5. When the WDT count overflows, the CPG starts supplying the clock and the processor resumes operation. The WOVF flag in WTCSR is not set when this happens. 6. Since the WDT continues counting from H'00, set the STBY bit in STBCR to 0 in the interrupt processing program and this will stop the WDT to count. When the STBY bit remains 1, the LSI again enters software standby mode when the WDT has counted up to H'80. This software standby mode can be canceled by a power-on reset.
Rev. 6.00 Jun. 12, 2007 Page 212 of 610 REJ09B0131-0600
Section 9 Watchdog Timer (WDT)
9.3.2
Changing Frequency
To change the multiplication ratio of PLL circuit 1, use the WDT. When changing the frequency only by switching the divider, do not use the WDT. 1. Before changing the frequency, always clear the TME bit in WTCSR to 0. When the TME bit is 1, an erroneous reset or interval timer interrupt may be generated when the count overflows. 2. Set the type of count clock used in the CKS2 to CKS0 bits in WTCSR and the initial values for the counter in WTCNT. These values should ensure that the time till count overflow is longer than the clock oscillation settling time. 3. When bits STC2 to STC0 in the frequency control register (FRQCR) is written, the processor stops temporarily. The WDT starts counting. 4. When the WDT count overflows, the CPG resumes supplying the clock and the processor resumes operation. The WOVF flag in WTCSR is not set when this happens. 5. WTCNT stops at the value of H'00. 6. Before changing WTCNT after the execution of the frequency change instruction, always confirm that the value of WTCNT is H'00 by reading WTCNT. 9.3.3 Using Watchdog Timer Mode
1. Set the WT/IT bit in WTCSR to 1, set the type of count clock in bits CKS2 to CKS0, and set the initial value of the counter in WTCNT. 2. Set the TME bit in WTCSR to 1 to start the count in watchdog timer mode. 3. While operating in watchdog timer mode, rewrite the counter periodically to H'00 to prevent the counter from overflowing. 4. When the counter overflows, the WDT sets the WOVF flag in WTCSR to 1 and generates a power-on reset. WTCNT then resumes counting.
Rev. 6.00 Jun. 12, 2007 Page 213 of 610 REJ09B0131-0600
Section 9 Watchdog Timer (WDT)
9.3.4
Using Interval Timer Mode
When operating in interval timer mode, interval timer interrupts are generated at every overflow of the counter. This enables interrupts to be generated at set periods. 1. Clear the WT/IT bit in WTCSR to 0, set the type of count clock in the CKS2 to CKS0 bits, and set the initial value of the counter in WTCNT. 2. Set the TME bit in WTCSR to 1 to start the count in interval timer mode. 3. When the WTCNT overflows, the WDT sets the IOVF flag in WTCSR to 1 and an interval timer interrupt request is sent to the INTC. The WTCNT then resumes counting.
9.4
Usage Note
Note the following when using the WDT. 1. When using the WDT in interval mode, no overflow occurs by the H'00 immediately after writing H'FF to WDTCNT. (IOVF in WTCSR is not set.) The overflow occurs at a point when the count reaches H'00 after one cycle. This does not occur when the WDT is used in watchdog timer mode.
Rev. 6.00 Jun. 12, 2007 Page 214 of 610 REJ09B0131-0600
Section 10 Power-Down Modes
Section 10 Power-Down Modes
This LSI supports the following power-down modes: sleep mode, software standby mode, module standby mode.
10.1
Features
* Supports sleep mode, software standby mode, and module standby 10.1.1 Types of Power-Down Modes
This LSI has the following power-down modes. * Sleep mode * Software standby mode * Module standby mode (cache, U-memory, UBC, H-UDI, and on-chip peripheral modules) Table 10.1 shows the methods to make a transition from the program execution state, as well as the CPU and peripheral module states in each mode and the procedures for canceling each mode.
Rev. 6.00 Jun. 12, 2007 Page 215 of 610 REJ09B0131-0600
Section 10 Power-Down Modes
Table 10.1 States of Power-Down Modes
State CPU Register Held On-Chip Memory Halts (contents remained) On-Chip Peripheral Modules Run Canceling Procedure * Interrupt other than user break * Software Execute SLEEP standby instruction with STBY bit in STBCR set to 1. Halts Halts Held Halts (contents remained) Halt Held * * Reset NMI, IRQ Reset
Mode Sleep
Transition Method Execute SLEEP instruction with STBY bit in STBCR cleared to 0.
CPG Runs
CPU Halts
Pins Held
Module Set MSTP bits in Runs standby STBCR2 to STBCR4 to 1.
Runs
Held
Specified Specified Held module halts module halts (contents remained)
*
Clear MSTP bit to 0
*
Power-on reset
Rev. 6.00 Jun. 12, 2007 Page 216 of 610 REJ09B0131-0600
Section 10 Power-Down Modes
10.2
Input/Output Pins
Table 10.2 lists the pins used for the power-down modes. Table 10.2 Pin Configuration
Pin Name Reset input pin Abbr. RES I/O Input Description Reset input signal. Reset by low level.
10.3
Register Descriptions
There are following registers used for the power-down modes. For details on the addresses of these registers and the states of these registers in each processing state, see section 20, List of Registers. * * * * Standby control register (STBCR) Standby control register 2 (STBCR2) Standby control register 3 (STBCR3) Standby control register 4 (STBCR4)
Rev. 6.00 Jun. 12, 2007 Page 217 of 610 REJ09B0131-0600
Section 10 Power-Down Modes
10.3.1
Standby Control Register (STBCR)
STBCR is an 8-bit readable/writable register that specifies the state of the power-down mode.
Bit 7 Bit Name STBY Initial Value 0 R/W R/W Description Standby Specifies transition to software standby mode. 0: Executing SLEEP instruction makes this LSI sleep mode 1: Executing SLEEP instruction makes this LSI software standby mode 6 to 4 All 0 R Reserved These bits are always read as 0. The write value should always be 0. 3 MDCHG 0 R/W MD2 to MD0 Pin Control Specifies whether or not the values of pins MD2 to MD0 are reflected in software standby mode. The values of pins MD2 to MD0 are reflected at returning from software standby mode by an interrupt when the MDCHG bit has been set to 1. 0: The values of pins MD2 to MO0 are not reflected in software standby mode. 1: The values of pins MD2 to MD0 are reflected in software standby mode. 2 to 0 All 0 R Reserved These bits are always read as 0. The write value should always be 0.
Rev. 6.00 Jun. 12, 2007 Page 218 of 610 REJ09B0131-0600
Section 10 Power-Down Modes
10.3.2
Standby Control Register 2 (STBCR2)
STBCR2 is an 8-bit readable/writable register that controls the operation of modules in powerdown mode.
Bit 7 Bit Name MSTP10 Initial Value 0 R/W R/W Description Module Stop Bit 10 When this bit is set to 1, the supply of the clock to the H-UDI is halted. 0: H-UDI operates 1: Clock supply to H-UDI halted 6 MSTP9 0 R/W Module Stop Bit 9 When this bit is set to 1, the supply of the clock to the UBC is halted. 0: UBC operates 1: Clock supply to UBC halted 5 to 3 All 0 R Reserved These bits are always read as 0. The write value should always be 0. 2 MSTP5 0 R/W Module Stop Bit 5 When this bit is set to 1, the supply of the clock to the cache memory is halted. 0: Cache memory operates 1: Clock supply to cache memory halted 1 MSTP4 0 R/W Module Stop Bit 4 When this bit is set to 1, the supply of the clock to the U memory is halted. 0: U memory operates 1: Clock supply to the U memory halted 0 0 R Reserved This bit is always read as 0. The write value should always be 0.
Rev. 6.00 Jun. 12, 2007 Page 219 of 610 REJ09B0131-0600
Section 10 Power-Down Modes
10.3.3
Standby Control Register 3 (STBCR3)
STBCR3 is an 8-bit readable/writable register that controls the operation of modules in powerdown mode.
Bit 7 to 5 Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 4 MSTP15 0 R/W Module Stop Bit 15 When this bit is set to 1, the supply of the clock to the CMT is halted. 0: CMT operates 1: Clock supply to CMT halted 3 0 R Reserved This bit is always read as 0. The write value should always be 0. 2 MSTP13 0 R/W Module Stop Bit 13 When this bit is set to 1, the supply of the clock to the SCIF2 is halted. 0: SCIF2 operates 1: Clock supply to SCIF2 halted 1 MSTP12 0 R/W Module Stop Bit 12 When this bit is set to 1, the supply of the clock to the SCIF1 is halted. 0: SCIF1 operates 1: Clock supply to SCIF1 halted 0 MSTP11 0 R/W Module Stop Bit 11 When this bit is set to 1, the supply of the clock to the SCIF0 is halted. 0: SCIF0 operates 1: Clock supply to SCIF0 halted
Rev. 6.00 Jun. 12, 2007 Page 220 of 610 REJ09B0131-0600
Section 10 Power-Down Modes
10.3.4
Standby Control Register 4 (STBCR4)
STBCR4 is an 8-bit readable/writable register that controls the operation of modules in powerdown mode.
Bit 7 to 5 Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 4 MSTP23 0 R/W Module Stop Bit 23 When this bit is set to 1, the supply of the clock to the HIF is halted. 0: HIF operates 1: Clock supply to HIF halted 3 to 1 All 0 R Reserved These bits are always read as 0. The write value should always be 0. 0 MSTP19 0 R/W Module Stop Bit 19 When this bit is set to 1, the supply of the clock to the EtherC and E-DMAC is halted. 0: EtherC and E-DMAC operate 1: Clock supply to EtherC and E-DMAC halted
Rev. 6.00 Jun. 12, 2007 Page 221 of 610 REJ09B0131-0600
Section 10 Power-Down Modes
10.4
10.4.1
Sleep Mode
Transition to Sleep Mode
Executing the SLEEP instruction when the STBY bit in STBCR is 0 causes a transition from the program execution state to sleep mode. Although the CPU halts immediately after executing the SLEEP instruction, the contents of its internal registers remain unchanged. The on-chip peripheral modules continue to operate in sleep mode and the clock continues to be output to the CKIO pin. 10.4.2 Canceling Sleep Mode
Sleep mode is canceled by an interrupt other than a user break (NMI, H-UDI, IRQ, and on-chip peripheral module) or a reset. Canceling with Interrupt: When a user-break, NMI, H-UDI, IRQ, or on-chip peripheral module interrupt occurs, sleep mode is canceled and interrupt exception handling is executed. When the priority level of an IRQ or on-chip peripheral module interrupt is lower than the interrupt mask level set in the status register (SR) of the CPU, an interrupt request is not accepted preventing sleep mode from being canceled. Canceling with Reset: Sleep mode is canceled by a power-on reset or an H-UDI reset.
Rev. 6.00 Jun. 12, 2007 Page 222 of 610 REJ09B0131-0600
Section 10 Power-Down Modes
10.5
10.5.1
Software Standby Mode
Transition to Software Standby Mode
This LSI switches from a program execution state to software standby mode by executing the SLEEP instruction when the STBY bit in STBCR is 1. In software standby mode, not only the CPU but also the clock and on-chip peripheral modules halt. The clock output from the CKIO pin also halts. The contents of the CPU and cache registers remain unchanged. Some registers of on-chip peripheral modules are, however, initialized. Table 10.3 lists the states of on-chip peripheral modules registers in software standby mode. Table 10.3 Register States in Software Standby Mode
Module Interrupt controller (INTC) Clock pulse generator (CPG) User break controller (UBC) Bus state controller (BSC) Ethernet controller (EtherC) Direct memory access controller for Ethernet controller (E-DMAC) I/O port User debugging interface (H-UDI) Serial communication interface with FIFO (SCIF0 to SCIF2) Compare match timer (CMT0 and CMT1) Host interface (HIF) Registers Initialized All registers Registers Retaining Data All registers All registers All registers All registers All registers All registers All registers All registers All registers All registers
Rev. 6.00 Jun. 12, 2007 Page 223 of 610 REJ09B0131-0600
Section 10 Power-Down Modes
The procedure for switching to software standby mode is as follows: 1. Clear the TME bit in the timer control register (WTCSR) of the WDT to 0 to stop the WDT. 2. Set the timer counter (WTCNT) of the WDT to 0 and bits CKS2 to CKS0 in WTCSR to appropriate values to secure the specified oscillation settling time. 3. After setting the STBY bit in STBCR to 1, execute the SLEEP instruction. 4. Software standby mode is entered and the clocks within this LSI are halted. 10.5.2 Canceling Software Standby Mode
Software standby mode is canceled by interrupts (NMI, IRQ) or a reset. Canceling with Interrupt: The WDT can be used for hot starts. When an NMI or IRQ interrupt is detected, the clock will be supplied to the entire LSI and software standby mode will be canceled after the time set in the timer control/status register of the WDT has elapsed. Interrupt exception handling is then executed. After the branch to the interrupt handling routine, clear the STBY bit in STBCR. WTCNT stops automatically. If the STBY bit is not cleared, WTCNT continues operation and a transition is made to software standby mode* when it reaches H'80. This function prevents data destruction due to the voltage rise by an unstable power supply voltage. IRQ cancels the software standby mode when the input condition matches the specified detect condition while the IRQn1S and IRQn0S bits in IRQCR are not B'00 (settings other than the low level detection). When the priority level of an IRQ interrupt is lower than the interrupt mask level set in the status register (SR) of the CPU, the execution of the instruction following the SLEEP instruction starts again after the cancellation of software standby mode. When the priority level of an IRQ interrupt is higher than the interrupt mask level set in the status register (SR) of the CPU, IRQ interrupt exception handling is executed after the cancellation of software standby mode. Note: * This software standby mode can be canceled only by a power-on reset.
Rev. 6.00 Jun. 12, 2007 Page 224 of 610 REJ09B0131-0600
Section 10 Power-Down Modes
Interrupt request Crystal oscillator settling time and PLL synchronization time WTCNT value
WDT overflow and branch to interrupt handling routine Clear bit STBCR.STBY before WTCNT reaches H'80. When STBCR.STBY is cleared, WTCNT halts automatically.
H'FF
H'80
Time
Figure 10.1 Canceling Standby Mode with STBY Bit in STBCR Canceling with Reset: Software standby mode is canceled by a power-on reset. Keep the RES pin low until the clock oscillation settles. The internal clock will continue to be output to the CKIO pin.
10.6
10.6.1
Module Standby Mode
Transition to Module Standby Mode
Setting the MSTP bits in the standby control registers (STBCR2 to STBCR4) to 1 halts the supply of clocks to the corresponding on-chip peripheral modules. This function can be used to reduce the power consumption in normal mode. In module standby mode, the states of the external pins of the on-chip peripheral modules change depending on the on-chip peripheral module and port settings. Almost all of the registers retains its previous state. 10.6.2 Canceling Module Standby Function
The module standby function can be canceled by clearing the MSTP bits in STBCR2 to STBCR4 to 0, or by a power-on reset.
Rev. 6.00 Jun. 12, 2007 Page 225 of 610 REJ09B0131-0600
Section 10 Power-Down Modes
Rev. 6.00 Jun. 12, 2007 Page 226 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
Section 11 Ethernet Controller (EtherC)
This LSI has an on-chip Ethernet controller (EtherC) conforming to the Ethernet or the IEEE802.3 MAC (Media Access Control) layer standard. Connecting a physical-layer LSI (PHY-LSI) complying with this standard enables the Ethernet controller (EtherC) to perform transmission and reception of Ethernet/IEEE802.3 frames. This LSI has one MAC layer interface. The Ethernet controller is connected to the direct memory access controller for Ethernet controller (E-DMAC) inside this LSI, and carries out high-speed data transfer to and from the memory. Figure 11.1 shows a configuration of the EtherC.
11.1
* * * * * *
Features
Transmission and reception of Ethernet/IEEE802.3 frames Supports 10/100 Mbps receive/transfer Supports full-duplex and half-duplex modes Conforms to IEEE802.3u standard MII (Media Independent Interface) Magic Packet detection and Wake-On-LAN (WOL) signal output Conforms to IEEE802.3x flow control
IFETH01C_000020030900
Rev. 6.00 Jun. 12, 2007 Page 227 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
E-DMAC
EtherC E-DMAC interface
MAC Transmit controller Receive controller
Command status interface
MII
PHY
Figure 11.1 Configuration of EtherC
Rev. 6.00 Jun. 12, 2007 Page 228 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
11.2
Input/Output Pins
Table 11.1 lists the pin configuration of the EtherC. Table 11.1 Pin Configuration
Port 0 Abbreviation I/O TX-CLK* Input Function Transmit Clock Timing reference signal for the TX-EN, MII_TXD3 to MII_TXD0, TX-ER signals 0 RX-CLK* Input Receive Clock Timing reference signal for the RX-DV, MII_RXD3 to MII_RXD0, RX-ER signals 0 TX-EN* Output Transmit Enable Indicates that transmit data is ready on pins MII_TXD3 to MII_TXD0. 0 0 0 MII_TXD3 to MII_TXD0* TX-ER* RX-DV* Output Output Input Transmit Data 4-bit transmit data Transmit Error Notifies the PHY-LSI of error during transmission Receive Data Valid Indicates that valid receive data is on pins MII_RXD3 to MII_RXD0. 0 0 0 0 0 MII_RXD3 to MII_RXD0* RX-ER* CRS COL MDC Input Input Input Input Output Receive Data 4-bit receive data Receive Error Identifies error state occurred during data reception. Carrier Detection Carrier detection signal Collision Detection Collision detection signal Management Data Clock Reference clock signal for information transfer via MDIO
Rev. 6.00 Jun. 12, 2007 Page 229 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
Port 0
Abbreviation I/O MDIO Input/ Output Input Output Output
Function Management Data I/O Bidirectional signal for exchange of management information between this LSI and PHY Link Status Inputs link status from PHY General-Purpose External Output Signal indicating value of register-bit (ECMR0-ELB) Wake-On-LAN Signal indicating reception of Magic Packet
0 0 0 Note:
LNKSTA EXOUT WOL *
MII signal conforming to IEEE802.3u
Rev. 6.00 Jun. 12, 2007 Page 230 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
11.3
Register Description
The EtherC has the following registers. For details on addresses and access sizes of registers, see section 20, List of Registers. MAC Layer Interface Control Register * EtherC mode register (ECMR) * EtherC status register (ECSR) * EtherC interrupt permission register (ECSIPR) * PHY interface register (PIR) * MAC address high register (MAHR) * MAC address low register (MALR) * Receive frame length register (RFLR) * PHY status register (PSR) * Transmit retry over counter register (TROCR) * Delayed collision detect counter register (CDCR) * Lost carrier counter register (LCCR) * Carrier not detect counter register (CNDCR) * CRC error frame counter register (CEFCR) * Frame receive error counter register (FRECR) * Too-short frame receive counter register (TSFRCR) * Too-long frame receive counter register (TLFRCR) * Residual-bit frame counter register (RFCR) * Multicast address frame counter register (MAFCR) * IPG register (IPGR) * Automatic PAUSE frame set register (APR) * Manual PAUSE frame set register (MPR) * Automatic PAUSE frame retransfer count set register (TPAUSER)
Rev. 6.00 Jun. 12, 2007 Page 231 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
11.3.1
EtherC Mode Register (ECMR)
ECMR is a 32-bit readable/writable register and specifies the operating mode of the Ethernet controller. The settings in this register are normally made in the initialization process following a reset. The operating mode setting must not be changed while the transmitting and receiving functions are enabled. To switch the operating mode, return the EtherC and E-DMAC to their initial states by means of the SWR bit in EDMR before making settings again.
Bit 31 to 20 Initial Bit Name Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 19 ZPF 0 R/W 0 time parameter PAUSE Frame Use Enable 0: Disables PAUSE frame control in which the TIME parameter is 0. The next frame is transmitted after the time indicated by the Timer value has elapsed. When the EtherC receives a PAUSE frame with the time indicated by the Timer value set to 0, the PAUSE frame is discarded. 1: Enables PAUSE frame control in which the TIME parameter is 0. A PAUSE frame with the Timer value set to 0 is transmitted when the number of data in the receive FIFO is less than the FCFTR value before the time indicated by the Timer value has not elapsed. When the EtherC receives a PAUSE frame with the time indicated by the Timer value set to 0, the transmit wait state is canceled. 18 PFR 0 R/W PAUSE Frame Receive Mode 0: PAUSE frame is not transferred to the E-DMAC 1: PAUSE frame is transferred to the E-DMAC 17 RXF 0 R/W Receive Flow Control Operating Mode 0: PAUSE frame detection function is disabled 1: Receive flow control function is enabled
Rev. 6.00 Jun. 12, 2007 Page 232 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
Bit 16
Initial Bit Name Value TXF 0
R/W R/W
Description Transmit Flow Control Operating mode 0: Transmit flow control function is disabled 1: Transmit flow control function is enabled
15 to 13
All 0
R
Reserved These bits are always read as 0. The write value should always be 0.
12
PRCEF
0
R/W
Permit Receive CRC Error Frame 0: A frame with a CRC error is received as a frame with an error. 1: A frame with a CRC error is received as a frame without an error. For a frame with an error, a CRC error is reflected in ECSR of the E-DMAC and the status of the receive descriptor. For a frame without an error, the frame is received as normal frame.
11, 10
All 0
R
Reserved These bits are always read as 0. The write value should always be 0.
9
MPDE
0
R/W
Magic Packet Detection Enable Enables or disables Magic Packet detection by hardware to allow activation from the Ethernet. 0: Magic Packet detection is not enabled 1: Magic Packet detection is enabled
8, 7
All 0
R
Reserved These bits are always read as 0. The write value should always be 0.
6
RE
0
R/W
Reception Enable If a frame is being received when this bit is switched from receive function enabled (RE = 1) to disabled (RE = 0), the receive function will be enabled until reception of the corresponding frame is completed. 0: Receive function is disabled 1: Receive function is enabled
Rev. 6.00 Jun. 12, 2007 Page 233 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
Bit 5
Initial Bit Name Value TE 0
R/W R/W
Description Transmission Enable If a frame is being transmitted when this bit is switched from transmit function enabled (TE = 1) to disabled (TE = 0), the transmit function will be enabled until transmission of the corresponding frame is completed. 0: Transmit function is disabled 1: Transmit function is enabled
4
0
R
Reserved This bit is always read as 0. The write value should always be 0.
3
ILB
0
R/W
Internal Loop Back Mode Specifies loopback mode in the EtherC. 0: Normal data transmission/reception is performed. 1: When DM = 1, data loopback is performed inside the MAC in the EtherC.
2
ELB
0
R/W
External Loop Back Mode This bit value is output directly to this LSI's generalpurpose external output pin (EXOUT). This bit is used for loopback mode directives, etc., in the LSI, using the EXOUT pin. In order for LSI loopback to be implemented using this function, the LSI must have a pin corresponding to the EXOUT pin. 0: Low-level output from the EXOUT pin 1: High-level output from the EXOUT pin
1
DM
0
R/W
Duplex Mode Specifies the EtherC transfer method. 0: Half-duplex transfer is specified 1: Full-duplex transfer is specified
Rev. 6.00 Jun. 12, 2007 Page 234 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
Bit 0
Initial Bit Name Value PRM 0
R/W R/W
Description Promiscuous Mode Setting this bit enables all Ethernet frames to be received. All Ethernet frames means all receivable frames, irrespective of differences or enabled/disabled status (destination address, broadcast address, multicast bit, etc.). 0: EtherC performs normal operation 1: EtherC performs promiscuous mode operation
11.3.2
EtherC Status Register (ECSR)
ECSR is a 32-bit readable/writable register and indicates the status in the EtherC. This status can be notified to the CPU by interrupts. When 1 is written to the PSRTO, LCHNG, MPD, and ICD, the corresponding flags can be cleared. Writing 0 does not affect the flag. For bits that generate interrupt, the interrupt can be enabled or disabled according to the corresponding bit in ECSIPR. The interrupts generated due to this status register are indicated in the ECI bit in EESR.
Bit Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 4 PSRTO 0 R/W PAUSE Frame Retransfer Retry Over Indicates that during the retransfer of PAUSE frames when the flow control is enabled, the number of retries has exceeded the upper limit set in the automatic PAUSE frame retransfer count set register (TPAUSER). 0: Number of PAUSE frame retransfers has not exceeded the upper limit 1: Number of PAUSE frame retransfers has exceeded the upper limit 3 0 R Reserved This bit is always read as 0. The write value should always be 0.
31 to 5
Rev. 6.00 Jun. 12, 2007 Page 235 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
Bit 2
Bit Name LCHNG
Initial Value 0
R/W R/W
Description Link Signal Change Indicates that the LNKSTA signal input from the PHY has changed from high to low or low to high. To check the current Link state, refer to the LMON bit in the PHY status register (PSR). 0: Changes in the LNKSTA signal are not detected 1: Changes in the LNKSTA signal are detected (high to low or low to high)
1
MPD
0
R/W
Magic Packet Detection Indicates that a Magic Packet has been detected on the line. 0: Magic Packet has not been detected 1: Magic Packet has been detected
0
ICD
0
R/W
Illegal Carrier Detection Indicates that the PHY has detected an illegal carrier on the line. If a change in the signal input from the PHY occurs before the software recognition period, the correct information may not be obtained. Refer to the timing specification for the PHY used. 0: LSI has not detected an illegal carrier on the line 1: LSI has detected an illegal carrier on the line
Rev. 6.00 Jun. 12, 2007 Page 236 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
11.3.3
EtherC Interrupt Permission Register (ECSIPR)
ECSIPR is a 32-bit readable/writable register that enables or disables the interrupt sources indicated by ECSR. Each bit can disable or enable interrupts corresponding to the bits in ECSR.
Bit Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 4 PSRTOIP 0 R/W PAUSE Frame Retransfer Retry Over Interrupt Enable 0: Interrupt notification by the PSRTO bit is disabled 1: Interrupt notification by the PSRTO bit is enabled 3 -- 0 R Reserved This bit is always read as 0. The write value should always be 0. 2 LCHNGIP 0 R/W LINK Signal Changed Interrupt Enable 0: Interrupt notification by the LCHNG bit is disabled 1: Interrupt notification by the LCHNG bit is enabled 1 MPDIP 0 R/W Magic Packet Detection Interrupt Enable 0: Interrupt notification by the MPD bit is disabled 1: Interrupt notification by the MPD bit is enabled 0 ICDIP 0 R/W Illegal Carrier Detection Interrupt Enable 0: Interrupt notification by the ICD bit is disabled 1: Interrupt notification by the ICD bit is enabled
31 to 5
Rev. 6.00 Jun. 12, 2007 Page 237 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
11.3.4
PHY Interface Register (PIR)
PIR is a 32-bit readable/writable register that provides a means of accessing the PHY registers via the MII.
Bit Bit Name Initial Value All 0 R/W Description R Reserved These bits are always read as 0. The write value should always be 0. 3 2 MDI MDO Undefined R 0 MII Management Data-In Indicates the level of the MDIO pin. R/W MII Management Data-Out Outputs the value set to this bit from the MDIO pin, when the MMD bit is 1. 1 MMD 0 R/W MII Management Mode Specifies the data read/write direction with respect to the MII. 0: Read direction is indicated 1: Write direction is indicated 0 MDC 0 R/W MII Management Data Clock Outputs the value set to this bit from the MDC pin and supplies the MII with the management data clock. For the method of accessing the MII registers, see section 11.4.4, Accessing MII Registers.
31 to 4
Rev. 6.00 Jun. 12, 2007 Page 238 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
11.3.5
MAC Address High Register (MAHR)
MAHR is a 32 -bit readable/writable register that specifies the upper 32 bits of the 48-bit MAC address. The settings in this register are normally made in the initialization process after a reset. The MAC address setting must not be changed while the transmitting and receiving functions are enabled. To switch the MAC address setting, return the EtherC and E-DMAC to their initial states by means of the SWR bit in EDMR before making settings again.
Bit Bit Name Initial Value R/W R/W Description MAC Address Bits These bits are used to set the upper 32 bits of the MAC address. If the MAC address is 01-23-45-67-89-AB (hexadecimal), the value set in this register is H'01234567.
31 to 0 MA47 to MA16 All 0
11.3.6
MAC Address Low Register (MALR)
MALR is a 32-bit readable/writable register that specifies the lower 16 bits of the 48-bit MAC address. The settings in this register are normally made in the initialization process after a reset. The MAC address setting must not be changed while the transmitting and receiving functions are enabled. To switch the MAC address setting, return the EtherC and E-DMAC to their initial states by means of the SWR bit in EDMR before making settings again.
Bit Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 15 to 0 MA15 to MA0 All 0 R/W MAC Address Bits 15 to 0 These bits are used to set the lower 16 bits of the MAC address. If the MAC address is 01-23-45-67-89-AB (hexadecimal), the value set in this register is H'000089AB.
31 to 16
Rev. 6.00 Jun. 12, 2007 Page 239 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
11.3.7
Receive Frame Length Register (RFLR)
RFLR is a 32-bit readable/writable register and it specifies the maximum frame length (in bytes) that can be received by this LSI. The settings in this register must not be changed while the receiving function is enabled.
Bit Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 11 to 0 RFL11 to RFL0 All 0 R/W Receive Frame Length 11 to 0 The frame length described here refers to all fields from the destination address up to and including the CRC data. Frame contents from the destination address up to and including the data are actually transferred to memory. CRC data is not included in the transfer. When data that exceeds the specified value is received, the part of the data that exceeds the specified value is discarded. H'000 to H'5EE: 1,518 bytes H'5EF: 1,519 bytes H'5F0: 1,520 bytes : : H'7FF: 2,047 bytes H'800 to H'FFF: 2,048 bytes
31 to 12
Rev. 6.00 Jun. 12, 2007 Page 240 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
11.3.8
PHY Status Register (PSR)
PSR is a read-only register that can read interface signals from the PHY.
Bit Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 0 LMON 0 R LNKSTA Pin Status The Link status can be read by connecting the Link signal output from the PHY to the LNKSTA pin. For the polarity, refer to the PHY specifications to be connected.
31 to 1
11.3.9
Transmit Retry Over Counter Register (TROCR)
TROCR is a 32-bit counter that indicates the number of frames that were unable to be transmitted in 16 transmission attempts including the retransfer. When 16 transmission attempts have failed, TROCR is incremented by 1. When the value in this register reaches H'FFFFFFFF, the count is halted. The counter value is cleared to 0 by a write to this register with any value.
Bit Bit Name Initial Value All 0 R/W R/W Description Transmit Retry Over Count These bits indicate the number of frames that were unable to be transmitted in 16 transmission attempts including the retransfer.
31 to 0 TROC31 to TROC0
Rev. 6.00 Jun. 12, 2007 Page 241 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
11.3.10 Delayed Collision Detect Counter Register (CDCR) CDCR is a 32-bit counter that indicates the number of delayed collisions on all lines from a start of transmission. When the value in this register reaches H'FFFFFFFF, count-up is halted. The counter value is cleared to 0 by a write to this register with any value.
Bit Bit Name Initial Value All 0 R/W R/W Description Delayed Collision Detect Count These bits indicate the number of delayed collisions on all lines from a start of transmission.
31 to 0 COSDC31 to COSDC0
11.3.11 Lost Carrier Counter Register (LCCR) LCCR is a 32-bit counter that indicates the number of times the carrier was lost during data transmission. When the value in this register reaches H'FFFFFFFF, the count is halted. The counter value is cleared to 0 by writing to this register with any value.
Bit Bit Name Initial Value All 0 R/W R/W Description Lost Carrier Count These bits indicate the number of times the carrier was lost during data transmission.
31 to 0 LCC31 to LCC0
11.3.12 Carrier Not Detect Counter Register (CNDCR) CNDCR is a 32-bit counter that indicates the number of times the carrier could not be detected while the preamble was being sent. When the value in this register reaches H'FFFFFFFF, the count is halted. The counter value is cleared to 0 by a write to this register with any value.
Bit Bit Name Initial Value All 0 R/W R/W Description Carrier Not Detect Count These bits indicate the number of times the carrier was not detected.
31 to 0 CNDC31 to CNDC0
Rev. 6.00 Jun. 12, 2007 Page 242 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
11.3.13 CRC Error Frame Counter Register (CEFCR) CEFCR is a 32-bit counter that indicates the number of times a frame with a CRC error was received. When the value in this register reaches H'FFFFFFFF, the count is halted. The counter value is cleared to 0 by a write to this register with any value.
Bit Bit Name Initial Value All 0 R/W R/W Description CRC Error Frame Count These bits indicate the count of CRC error frames received.
31 to 0 CEFC31 to CEFC0
11.3.14 Frame Receive Error Counter Register (FRECR) FRECR is a 32-bit counter that indicates the number of frames input from the PHY for which a receive error was indicated by the RX-ER pin. FRECR is incremented each time the RX-ER pin becomes active. When the value in this register reaches H'FFFFFFFF, the count is halted. The counter value is cleared to 0 by a write to this register with any value.
Bit Bit Name Initial Value All 0 R/W R/W Description Frame Receive Error Count These bits indicate the count of errors during frame reception.
31 to 0 FREC31 to FREC0
11.3.15 Too-Short Frame Receive Counter Register (TSFRCR) TSFRCR is a 32-bit counter that indicates the number of frames of fewer than 64 bytes that have been received. When the value in this register reaches H'FFFFFFFF, the count is halted. The counter value is cleared to 0 by a write to this register with any value.
Bit Bit Name Initial Value All 0 R/W R/W Description Too-Short Frame Receive Count These bits indicate the count of frames received with a length of less than 64 bytes.
31 to 0 TSFC31 to TSFC0
Rev. 6.00 Jun. 12, 2007 Page 243 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
11.3.16 Too-Long Frame Receive Counter Register (TLFRCR) TLFRCR is a 32-bit counter that indicates the number of frames received with a length exceeding the value specified by the receive frame length register (RFLR). When the value in this register reaches H'FFFFFFFF, the count is halted. TLFRCR is not incremented when a frame containing residual bits is received. In this case, the reception of the frame is indicated in the residual-bit frame counter register (RFCR). The counter value is cleared to 0 by a write to this register with any value.
Bit Bit Name Initial Value All 0 R/W R/W Description Too-Long Frame Receive Count These bits indicate the count of frames received with a length exceeding the value in RFLR.
31 to 0 TLFC31 to TLFC0
11.3.17 Residual-Bit Frame Counter Register (RFCR) RFCR is a 32-bit counter that indicates the number of frames received containing residual bits (less than an 8-bit unit). When the value in this register reaches H'FFFFFFFF, the count is halted. The counter value is cleared to 0 by a write to this register with any value.
Bit Bit Name Initial Value All 0 R/W R/W Description Residual-Bit Frame Count These bits indicate the count of frames received containing residual bits.
31 to 0 RFC31 to RFC0
11.3.18 Multicast Address Frame Counter Register (MAFCR) MAFCR is a 32-bit counter that indicates the number of frames received with a specified multicast address. When the value in this register reaches H'FFFFFFFF, the count is halted. The counter value is cleared to 0 by a write to this register with any value.
Bit Bit Name Initial Value All 0 R/W R/W Description Multicast Address Frame Count These bits indicate the count of multicast frames received.
31 to 0 MAFC31 to MAFC0
Rev. 6.00 Jun. 12, 2007 Page 244 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
11.3.19 IPG Register (IPGR) IPGR sets the IPG (Inter Packet Gap). This register must not be changed while the transmitting and receiving functions of the EtherC mode register (ECMR) are enabled. (For details, refer to section 11.4.6, Operation by IPG Setting.)
Bit Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 4 to 0 IPG4 to IPG0 H'13 R/W Inter Packet Gap Sets the IPG value every 4-bit time. H'00: 20-bit time H'01: 24-bit time : : : : H'13: 96-bit time (Initial value) H'1F: 144-bit time
31 to 5
11.3.20 Automatic PAUSE Frame Set Register (APR) APR sets the TIME parameter value of the automatic PAUSE frame. When transmitting the automatic PAUSE frame, the value set in this register is used as the TIME parameter of the PAUSE frame.
Bit Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 15 to 0 AP15 to AP0 All 0 R/W Automatic PAUSE Sets the TIME parameter value of the automatic PAUSE frame. At this time, 1 bit means 512-bit time.
31 to 16
Rev. 6.00 Jun. 12, 2007 Page 245 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
11.3.21 Manual PAUSE Frame Set Register (MPR) MPR sets the TIME parameter value of the manual PAUSE frame. When transmitting the manual PAUSE frame, the value set to this register is used as the TIME parameter of the PAUSE frame.
Bit Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 15 to 0 MP15 to MP0 All 0 R/W Manual PAUSE Sets the TIME parameter value of the manual PAUSE frame. At this time, 1 bit means 512-bit time. Read values are undefined.
31 to 16
11.3.22 Automatic PAUSE Frame Retransfer Count Set Register (TPAUSER) TPAUSER sets the upper limit of the number of times of the PAUSE frame retransfer. TPAUSER must not be changed while the transmitting function is enabled.
Bit 31 to 16 Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 15 to 0 TPAUSE15 All 0 to TPAUSE0 R/W Upper Limit of the Number of Times of PAUSE Frame Retransfer H'0000: Unlimited number of times of retransfer H'0001: Retransfer once : : H'FFFF: Number of times of retransfer is 65535
Rev. 6.00 Jun. 12, 2007 Page 246 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
11.4
Operation
The overview of the Ethernet controller (EtherC) are shown below. The EtherC transmits and receives PAUSE frames conforming to the Ethernet/IEEE802.3 frames. 11.4.1 Transmission
The EtherC transmitter assembles the transmit data on the frame and outputs to MII when there is a transmit request from the E-DMAC. The data transmitted via the MII is transmitted to the lines by PHY-LSI. Figure 11.3 shows the state transition of the EtherC transmitter.
Rev. 6.00 Jun. 12, 2007 Page 247 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
TE set Idle
FDPX
Start of transmission (preamble transmission) Carrier non-detection Retransfer initiation Collision Carrier detection HDPX FDPX
Transmission halted
HDPX TE reset
Carrier detection
Carrier detection Reset
Retransfer processing*1 Failure of 15 retransfer attempts or collision after 512-bit time
Carrier non-detection Collision SFD transmission Error Collision*2
Error detection Error notification
Error
Data transmission Collision*2
Error [Legend] CRC FDPX: Full Duplex Normal transmission transmission HDPX: Half Duplex SFD: Start Frame Delimiter Notes: 1. Transmission retry processing includes both jam transmission that depends on collision detection and the adjustment of transmission intervals based on the back-off algorithm. 2. Transmission is retried only when data of 512 bits or less (including the preamble and SFD) is transmitted. When a collision is detected during the transmission of data greater than 512 bits, only jam is transmitted and transmission based on the back-off algorithm is not retried.
Figure 11.2 EtherC Transmitter State Transitions 1. When the transmit enable (TE) bit is set, the transmitter enters the transmit idle state. 2. When a transmit request is issued by the transmit E-DMAC, the EtherC sends the preamble after a transmission delay equivalent to the frame interval time. If full-duplex transfer is selected, which does not require carrier detection, the preamble is sent as soon as a transmit request is issued by the E-DMAC.
Rev. 6.00 Jun. 12, 2007 Page 248 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
3. The transmitter sends the SFD, data, and CRC sequentially. At the end of transmission, the transmit E-DMAC generates a transmission complete interrupt (TC). If a collision or the carrier-not-detected state occurs during data transmission, these are reported as interrupt sources. 4. After waiting for the frame interval time, the transmitter enters the idle state, and if there is more transmit data, continues transmitting. 11.4.2 Reception
The EtherC receiver separates the frame data (MII into preamble, SFD, DA (destination address), SA (Source address), type/length, Data, and CRC data) and outputs DA, SA, type/length, Data to the E-DMAC. Figure 11.3 shows the state transitions of the EtherC receiver.
Illegal carrier detection RX-DV negation
Idle RE set Preamble detection
Start of frame reception
Wait for SFD reception SFD reception Destination address reception Own destination address or broadcast or multicast or promiscuous Data reception End of reception CRC reception
Reception halted
RE reset Promiscuous and other station destination address
Reset
Error notification*
Error detection
Receive error detection
Receive error detection
Normal reception [Legend] SFD: Start frame delimiter Note: * The error frame also transmits data to the buffer.
Figure 11.3 EtherC Receiver State Transmissions
Rev. 6.00 Jun. 12, 2007 Page 249 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
1. When the receive enable (RE) bit is set, the receiver enters the receive idle state. 2. When an SFD (start frame delimiter) is detected after a receive packet preamble, the receiver starts receive processing. Discards a frame with an invalid pattern. 3. In normal mode, if the destination address matches the receiver's own address, or if broadcast or multicast transmission or promiscuous mode is specified, the receiver starts data reception. 4. Following data reception from the MII, the receiver carries out a CRC check. The result is indicated as a status bit in the descriptor after the frame data has been written to memory. Reports an error status in the case of an abnormality. 5. After one frame has been received, if the receive enable bit is set (RE = 1) in the EtherC mode register, the receiver prepares to receive the next frame. 11.4.3 MII Frame Timing
Each MII Frame timing is shown in figure 11.4.
TX-CLK TX-EN TXD3 to TXD0 TX-ER CRS COL Preamble SFD Data CRC
Figure 11.4 (1) MII Frame Transmit Timing (Normal Transmission)
TX-CLK TX-EN MII_TXD3 to MII_TXD0 TX-ER CRS COL Preamble JAM
Figure 11.4 (2) MII Frame Transmit Timing (Collision)
Rev. 6.00 Jun. 12, 2007 Page 250 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
TX-CLK TX-EN MII_TXD3 to MII_TXD0 TX-ER CRS COL Preamble SFD Data
Figure 11.4 (3) MII Frame Transmit Timing (Transmit Error)
RX-CLK RX-DV MII_RXD3 to MII_RXD0 RX-ER Preamble SFD Data CRC
Figure 11.4 (4) MII Frame Receive Timing (Normal Reception)
RX-CLK RX-DV MII_RXD3 to MII_RXD0 RX-ER Preamble SFD Data XXXX
Figure 11.4 (5) MII Frame Receive Timing (Reception Error (1))
RX-CLK RX-DV MII_RXD3 to MII_RXD0 RX-ER XXXX 1110 XXXX
Figure 11.4 (6) MII Fame Receive Timing (Reception Error (2))
Rev. 6.00 Jun. 12, 2007 Page 251 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
11.4.4
Accessing MII Registers
MII registers in the PHY are accessed via this LSI's PHY interface register (PIR). Connection is made as a serial interface in accordance with the MII frame format specified in IEEE802.3u. MII Management Frame Format: The format of an MII management frame is shown in figure 11.8. To access an MII register, a management frame is implemented by the program in accordance with the procedures shown in MII Register Access Procedure.
Access Type Item Number of bits Read Write [Legend] PRE: ST: OP: PHYAD: PRE 32 1..1 1..1 ST 2 01 01 OP 2 10 01 MII Management Frame PHYAD 5 00001 00001 REGAD 5 RRRRR RRRRR TA 2 Z0 10 DATA 16 D..D D..D X IDLE
32 consecutive 1s Write of 01 indicating start of frame Write of code indicating access type Write of 0001 if the PHY address is 1 (sequential write starting with the MSB). This bit changes depending on the PHY address. REGAD: Write of 0001 if the register address is 1 (sequential write starting with the MSB). This bit changes depending on the PHY register address. TA: Time for switching data transmission source on MII interface (a) Write: 10 written (b) Read: Bus release (notation: Z0) performed DATA: 16-bit data. Sequential write or read from MSB (a) Write: 16-bit data write (b) Read: 16-bit data read IDLE: Wait time until next MII management format input (a) Write: Independent bus release (notation: X) performed (b) Read: Bus already released in TA; control unnecessary
Figure 11.5 MII Management Frame Format
Rev. 6.00 Jun. 12, 2007 Page 252 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
MII Register Access Procedure: The program accesses MII registers via the PHY interface register (PIR). Access is implemented by a combination of 1-bit-unit data write, 1-bit-unit data read, bus release, and independent bus release. Figure 11.9 shows the MII register access timing. The timing will differ depending on the PHY type.
(1) Write to PHY interface register MMD = 1 MDO = write data MDC = 0 MDC MDO
(2) Write to PHY interface register MMD = 1 MDO = write data MDC = 1
(1) (2)
(3)
1-bit data write timing relationship
(3) Write to PHY interface register MMD = 1 MDO = write data MDC = 0
Figure 11.6 (1) 1-Bit Data Write Flowchart
Rev. 6.00 Jun. 12, 2007 Page 253 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
(1)
Write to PHY interface register MMD = 0 MDC = 0 MDC MDO
(2)
Write to PHY interface register (1) (2) MMD = 0 MDC = 1 (3) Bus release timing relationship
(3)
Write to PHY interface register MMD = 0 MDC = 0
Figure 11.6 (2) Bus Release Flowchart (TA in Read in Figure 11.5)
(1) Write to PHY interface register MMD = 0 MDC = 1 MDC MDI (2) Read from PHY interface register read MMD = 0 MMC = 1 MDI is read data (1) (2) (3) 1-bit data read timing relationship
(3) Write to PHY interface register MMD = 0 MDC = 0
Figure 11.6 (3) 1-Bit Data Read Flowchart
Rev. 6.00 Jun. 12, 2007 Page 254 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
(1) Write to PHY interface register MMD = 0 MDC = 0 MDC MDO
(1) Independent bus release timing relationship
Figure 11.6 (4) Independent Bus Release Flowchart (IDLE in Write in Figure 11.5) 11.4.5 Magic Packet Detection
The EtherC has a Magic Packet detection function. This function provides a Wake-On-LAN (WOL) facility that activates various peripheral devices connected to a LAN from the host device or other source. This makes it possible to construct a system in which a peripheral device receives a Magic Packet sent from the host device or other source, and activates itself. When the Magic Packet is detected, data is stored in the FIFO of the E-DMAC by the broadcast packet that has received data previously and the EtherC is notified of the receiving status. To return to normal operation from the interrupt processing, initialize the EtherC and E-DMAC by using the SWR bit in the E-DMAC mode register (EDMR). With a Magic Packet, reception is performed regardless of the destination address. As a result, this function is valid, and the WOL pin enabled, only in the case of a match with the destination address specified by the format in the Magic Packet. Further information on Magic Packets can be found in the technical documentation published by AMD Corporation. The procedure for using the WOL function with this LSI is as follows. 1. Disable interrupt source output by means of the various interrupt enable/mask registers. 2. Set the Magic Packet detection enable bit (MPDE) in the EtherC mode register (ECMR). 3. Set the Magic Packet detection interrupt enable bit (MPDIP) in the EtherC interrupt enable register (ECSIPR) to the enable setting. 4. If necessary, set the CPU operating mode to sleep mode or set supporting functions to module standby mode. 5. When a Magic Packet is detected, an interrupt is sent to the CPU. The WOL pin notifies peripheral LSIs that the Magic Packet has been detected.
Rev. 6.00 Jun. 12, 2007 Page 255 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
11.4.6
Operation by IPG Setting
The EtherC has a function to change the non-transmission period IPG (Inter Packet Gap) between transmit frames. By changing the set values of the IPG setting register (IPGR), the transmission efficiency can be raised and lowered from the standard value. IPG settings are prescribed in IEEE802.3 standards. When changing settings, adequately check that the respective devices can operate smoothly on the same network.
Case A (short IPG)
[1]
[2]
[3]
[4]
[5]
......
Packet Case B (long IPG)
IPG*
[1]
[2]
[3]
[4]
......
Note: * IPG may be longer than the set value, depending on the state of the circuit and the system bus.
Figure 11.7 Changing IPG and Transmission Efficiency 11.4.7 Flow Control
The EtherC supports flow control functions conforming to IEEE802.3x in full-duplex operations. Flow control can be applied to both receive and transmit operations. The methods for transmitting PAUSE frames when controlling flow are as follows: Automatic PAUSE Frame Transmission: For receive frames, PAUSE frames are automatically transmitted when the number of data in the receive FIFO (included in E-DMAC) reaches the value set in the flow control FIFO threshold register (FCFTR) of the E-DMAC. The TIME parameter included in the PAUSE frame at this time is set by the automatic PAUSE frame setting register (APR). The automatic PAUSE frame transmission is repeated until the number of data in the receive FIFO becomes less than the FCFTR setting as the receive data is read from the FIFO. The upper limit of the number of retransfers of the PAUSE frame can also be set by the automatic PAUSE frame retransfer count set register (TPAUSER). In this case, PAUSE frame transmission is repeated until the number of data becomes FCFTR value set or below, or the number of transmits reaches the value set by TPAUSER. The automatic PAUSE frame transmission is enabled when the TXF bit in the EtherC mode register (ECMR) is 1.
Rev. 6.00 Jun. 12, 2007 Page 256 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
Manual PAUSE Frame Transmission: PAUSE frames are transmitted by directives from the software. When writing the Timer value to the manual PAUSE frame set register (MPR), manual PAUSE frame transmission is started. With this method, PAUSE frame transmission is carried out only once. PAUSE Frame Reception: The next frame is not transmitted until the time indicated by the Timer value elapses after receiving a PAUSE frame. However, the transmission of the current frame is continued. A received PAUSE frame is valid only when the RXF bit in the EtherC mode register (ECMR) is set to 1.
11.5
Connection to PHY-LSI
Figure 11.8 shows an example of connection to a DP83846AVHG (National Semiconductor Corporation).
MII (Media independent interface) DP83846AVHG TX_ER TXD3 TXD2 TXD1 TXD0 TX_EN TX_CLK MDC MDIO RXD3 RXD2 RXD1 RXD0 RX_CLK CRS COL RX_DV RX_ER
This LSI
TX-ER MII_TXD3 MII_TXD2 MII_TXD1 MII_TXD0 TX-EN TX-CLK MDC MDIO MII_RXD3 MII_RXD2 MII_RXD1 MII_RXD0 RX-CLK CRS COL RX-DV RX-ER
Figure 11.8 Example of Connection to DP83846AVHG
Rev. 6.00 Jun. 12, 2007 Page 257 of 610 REJ09B0131-0600
Section 11 Ethernet Controller (EtherC)
11.6
Usage Notes
* Conditions for Setting LCHNG Bit Even if the level of the signal input to the LNKSTA pin is not changed, the LCHNG bit in ECSR may be set. It may happen when the pin function is changed from port to LNKSTA by PCCRH2 of the PFC or when a software reset caused by the SWR bit in EDMR is cleared while the LNKSTA pin is being driven high. This is because the LNKSTA signal is internally fixed low when the pin functions as a port or during the software reset state regardless of the external pin level. Clear the LCHNG bit before setting the LCHNGIP bit in ECSIPR not to request a LINK signal changed interrupt accidentally. * Flow Control Defect 1 Once a PAUSE frame is received while the receiving flow control is enabled in full-duplex mode (the RXF bit in ECMR = 1), each time when the local station receives a normal unicast frame (non-PAUSE frame without a CRC error), the TIME parameter specified by the PAUSE frame that has been previously received is incorrectly applied. As a result, unnecessary waiting time is generated to slow down the transmission throughput. The TIME parameter value is maintained until another PAUSE frame is received. This defect can be prevented if the destination station supports the function to transmit the 0 time PAUSE frame as the same as this LSI does. Enable the use of 0 time PAUSE frame in this LSI (the ZPF bit in ECMR = 1) before the 0 time PAUSE frame is received from the destination station. This clears the TIME parameter incorrectly maintained in the EtherC and prevents the unnecessary waiting time for transmission to be generated. * Flow Control Defect 2 When a PAUSE period is generated while the transmitting/receiving flow control is enabled in full-duplex mode (the TXF/RXF bit in ECMR = 1), non-PAUSE frames are waited for transmission (this is a normal operation) whereas PAUSE frames are incorrectly waited for transmission. The transmission of non-PAUSE frames in a PAUSE period is prohibited, though the transmission of PAUSE frames is enabled in IEEE802.3. When a PAUSE period is generated by the request from the destination station (that is, a PAUSE frame is received from the destination station), the load of the destination station is high and that of the local station is not so high. Therefore, the transmission of PAUSE frames during this period is less likely to happen. The ratio that this defect actually affects the operation in this LSI is rather low.
Rev. 6.00 Jun. 12, 2007 Page 258 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
This LSI includes a direct memory access controller (E-DMAC) directly connected to the Ethernet controller (EtherC). A large proportion of buffer management is controlled by the E-DMAC itself using descriptors. This lightens the load on the CPU and enables efficient data transfer control to be achieved. Figure 12.1 shows the configuration of the E-DMAC, and the descriptors and transmit/receive buffers in memory.
12.1
Features
The E-DMAC has the following features: * * * * The load on the CPU is reduced by means of a descriptor management system Transmit/receive frame status information is indicated in descriptors Achieves efficient system bus utilization through the use of block transfer (16-byte units) Supports single-frame/multi-buffer operation
This LSI Internal bus Transmit buffer Transmit descriptor External bus interface
E-DMAC
Receive buffer Receive descriptor
Internal bus interface
Descriptor information Transmit DMAC Descriptor information Receive DMAC
Transmit FIFO
Receive FIFO
EtherC
External memory
Figure 12.1 Configuration of E-DMAC, and Descriptors and Buffers
EDMAS20C_000020030900
Rev. 6.00 Jun. 12, 2007 Page 259 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
12.2
Register Descriptions
The E-DMAC has the following registers. For addresses and access sizes of these registers, see section 20, List of Registers. * * * * * * * * * * * * * * * * * * * E-DMAC mode register (EDMR) E-DMAC transmit request register (EDTRR) E-DMAC receive request register (EDRRR) Transmit descriptor list address register (TDLAR) Receive descriptor list address register (RDLAR) EtherC/E-DMAC status register (EESR) EtherC/E-DMAC status interrupt permission register (EESIPR) Transmit/receive status copy enable register (TRSCER) Receive missed-frame counter register (RMFCR) Transmit FIFO threshold register (TFTR) FIFO depth register (FDR) Receiving method control register (RMCR) E-DMAC operation control register (EDOCR) Receive buffer write address register (RBWAR) Receive descriptor fetch address register (RDFAR) Transmit buffer read address register (TBRAR) Transmit descriptor fetch address register (TDFAR) Flow control FIFO threshold register (FCFTR) Transmit interrupt register (TRIMD)
Rev. 6.00 Jun. 12, 2007 Page 260 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
12.2.1
E-DMAC Mode Register (EDMR)
EDMR is a 32-bit readable/writable register that specifies the operating mode of the E-DMAC. The settings in this register are normally made in the initialization process following a reset. If the EtherC and E-DMAC are initialized by means of this register during data transmission, abnormal data may be sent onto the line. Operating mode settings must not be changed while the transmit and receive functions are enabled. To change the operating mode, the EtherC and E-DMAC modules are got into at their initial state by means of the software reset bit (SWR) in this register, then make new settings. It takes 64 cycles of the internal bus clock B to initialize the EtherC and E-DMAC. Therefore, registers of the EtherC and E-DMAC should be accessed after 64 cycles of the internal bus clock B has elapsed.
Bit 31 to 7 Bit Name Initial value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 6 DE 0 R/W E-DMAC Data Endian Convert Selects whether or not the endian format is converted on data transfer by the E-DMAC. However, the endian format of the descriptors and E-DMAC register values are not converted regardless of this bit setting. 0: Endian format not converted (big endian) 1: Endian format converted (little endian) 5 4 DL1 DL0 0 0 R/W R/W Descriptor Length These bits specify the descriptor length. 00: 16 bytes 01: 32 bytes 10: 64 bytes 11: Reserved (setting prohibited) 3 to 1 All 0 R Reserved These bits are always read as 0. The write value should always be 0.
Rev. 6.00 Jun. 12, 2007 Page 261 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
Bit 0
Bit Name SWR
Initial value 0
R/W R/W
Description Software Reset Writing 1 in this bit initializes registers of the E-DMAC other than TDLAR, RDLAR, and RMFCR and registers of the EtherC. While a software reset is issued (64 cycles of the internal bus clock B), accesses to the all Ethernet-related registers are prohibited. Software reset period (example): When B = 50 MHz: 1.28 S When B = 33 MHz: 1.94 S This bit is always read as 0. 0: Writing 0 is ignored (E-DMAC operation is not affected) 1: Writing 1 resets the EtherC and E-DMAC and then automatically cleared
12.2.2
E-DMAC Transmit Request Register (EDTRR)
The EDTRR is a 32-bit readable/writable register that issues transmit directives to the E-DMAC. When transmission of one frame is completed, the next descriptor is read. If the transmit descriptor active bit in this descriptor has the "active" setting, transmission is continued. If the transmit descriptor active bit has the "inactive" setting, the TR bit is cleared and operation of the transmit DMAC is halted.
Bit 31 to 1 Bit Name Initial value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 0 TR 0 R/W Transmit Request 0: Transmission-halted state. Writing 0 does not stop transmission. Termination of transmission is controlled by the active bit in the transmit descriptor 1: Start of transmission. The relevant descriptor is read and a frame is sent with the transmit active bit set to 1
Rev. 6.00 Jun. 12, 2007 Page 262 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
12.2.3
E-DMAC Receive Request Register (EDRRR)
EDRRR is a 32-bit readable/writable register that issues receive directives to the E-DMAC. When the receive request bit is set, the E-DMAC reads the relevant receive descriptor. If the receive descriptor active bit in the descriptor has the "active" setting, the E-DMAC prepares for a receive request from the EtherC. When one receive buffer of data has been received, the E-DMAC reads the next descriptor and prepares to receive the next frame. If the receive descriptor active bit in the descriptor has the "inactive" setting, the RR bit is cleared and operation of the receive DMAC is halted.
Bit 31 to 1 Bit Name Initial value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 0 RR 0 R/W Receive Request 0: The receive function is disabled* 1: A receive descriptor is read and the E-DMAC is ready to receive Note: * If the receive function is disabled during frame reception, write-back is not performed successfully to the receive descriptor. Following pointers to read a receive descriptor become abnormal and the E-DMAC cannot operate successfully. In this case, to make the E-DMAC reception enabled again, execute a software reset by the SWR bit in EDMR. To make the E-DMAC reception disabled without executing a software reset, set the RE bit in ECMR. Next, after the E_DMAC has completed the reception and write-back to the receive descriptor has been confirmed, disable the receive function of this register.
Rev. 6.00 Jun. 12, 2007 Page 263 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
12.2.4
Transmit Descriptor List Address Register (TDLAR)
TDLAR is a 32-bit readable/writable register that specifies the start address of the transmit descriptor list. Descriptors have a boundary configuration in accordance with the descriptor length indicated by the DL bit in EDMR. This register must not be written to during transmission. Modifications to this register should only be made while transmission is disabled by the TR bit (= 0) in the E-DMAC transmit request register (EDTRR).
Bit 31 to 0 Bit Name TDLA31 to TDLA0 Initial value All 0 R/W R/W Description Transmit Descriptor Start Address The lower bits are set as follows according to the specified descriptor length. 16-byte boundary: TDLA3 to TDLA0 = 0000 32-byte boundary: TDLA4 to TDLA0 = 00000 64-byte boundary: TDLA5 to TDLA0 = 000000
12.2.5
Receive Descriptor List Address Register (RDLAR)
RDLAR is a 32-bit readable/writable register that specifies the start address of the receive descriptor list. Descriptors have a boundary configuration in accordance with the descriptor length indicated by the DL bit in EDMR. This register must not be written to during reception. Modifications to this register should only be made while reception is disabled by the RR bit (= 0) in the E-DMAC Receive Request Register (EDRRR).
Bit 31 to 0 Bit Name RDLA31 to RDLA0 Initial value All 0 R/W R/W Description Receive Descriptor Start Address The lower bits are set as follows according to the specified descriptor length. 16-byte boundary: RDLA3 to RDLA0 = 0000 32-byte boundary: RDLA4 to RDLA0 = 00000 64-byte boundary: RDLA5 to RDLA0 = 000000
Rev. 6.00 Jun. 12, 2007 Page 264 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
12.2.6
EtherC/E-DMAC Status Register (EESR)
EESR is a 32-bit readable/writable register that shows communications status information on the E-DMAC in combination with the EtherC. The information in this register is reported in the form of interrupts. Individual bits are cleared by writing 1 (however, bit 22 (ECI) is a read-only bit and not to be cleared by writing 1) and are not affected by writing 0. Each interrupt source can also be masked by means of the corresponding bit in the EtherC/E-DMAC status interrupt permission register (EESIPR). The interrupts generated by this register are EINT0. For interrupt priority, see section 6.5, Interrupt Exception Handling Vector Table.
Bit 31 Bit Name Initial value 0 R/W R Description Reserved This bit is always read as 0. The write value should always be 0. 30 TWB 0 R/W Write-Back Complete Indicates that write-back from the E-DMAC to the corresponding descriptor has completed. This operation is enabled when the TIS bit in TRIMD is set to 1. 0: Write-back has not completed, or no transmission directive 1: Write-back has completed 29 to 27 All 0 R Reserved These bits are always read as 0. The write value should always be 0. 26 TABT 0 R/W Transmit Abort Detection Indicates that the EtherC aborts transmitting a frame because of failures during transmitting the frame. 0: Frame transmission has not been aborted or no transmit directive 1: Frame transmit has been aborted
Rev. 6.00 Jun. 12, 2007 Page 265 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
Bit 25
Bit Name RABT
Initial value 0
R/W R/W
Description Receive Abort Detection Indicates that the EtherC aborts receiving a frame because of failures during receiving the frame. 0: Frame reception has not been aborted or no receive directive 1: Frame receive has been aborted
24
RFCOF
0
R/W
Receive Frame Counter Overflow Indicates that the receive FIFO frame counter has overflowed. 0: Receive frame counter has not overflowed 1: Receive frame counter overflows
23
ADE
0
R/W
Address Error Indicates that the memory address that the E-DMAC tried to transfer is found illegal. 0: Illegal memory address not detected (normal operation) 1: Illegal memory address detected Note: When an address error is detected, the E-DMAC halts transmitting/receiving. To resume the operation, set the E-DMAC again after software reset by means of the SWR bit in EDMR.
22
ECI
0
R
EtherC Status Register Interrupt Source This bit is a read-only bit. When the source of an ECSR interrupt in the EtherC is cleared, this bit is also cleared. 0: EtherC status interrupt source has not been detected 1: EtherC status interrupt source has been detected
Rev. 6.00 Jun. 12, 2007 Page 266 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
Bit 21
Bit Name TC
Initial value 0
R/W R/W
Description Frame Transmit Complete Indicates that all the data specified by the transmit descriptor has been transmitted to the EtherC. The transfer status is written back to the relevant descriptor. When 1-frame transmission is completed for 1-frame/1-buffer processing, or when the last data in the frame is transmitted and the transmission descriptor valid bit (TACT) in the next descriptor is not set for multiple-frame buffer processing, transmission is completed and this bit is set to 1. After frame transmission, the E-DMAC writes the transmission status back to the descriptor. 0: Transfer not complete, or no transfer directive 1: Transfer complete
20
TDE
0
R/W
Transmit Descriptor Empty Indicates that the transmission descriptor valid bit (TACT) in the descriptor is not set when the E-DMAC reads the transmission descriptor when the previous descriptor is not the last one of the frame for multiplebuffer frame processing. As a result, an incomplete frame may be transmitted. 0: Transmit descriptor active bit TACT = 1 detected 1: Transmit descriptor active bit TACT = 0 detected When transmission descriptor empty (TDE = 1) occurs, execute a software reset and initiate transmission. In this case, the address that is stored in the transmit descriptor list address register (TDLAR) is transmitted first.
19
TFUF
0
R/W
Transmit FIFO Underflow Indicates that underflow has occurred in the transmit FIFO during frame transmission. Incomplete data is sent onto the line. 0: Underflow has not occurred 1: Underflow has occurred
Rev. 6.00 Jun. 12, 2007 Page 267 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
Bit 18
Bit Name FR
Initial value 0
R/W R/W
Description Frame Reception Indicates that a frame has been received and the receive descriptor has been updated. This bit is set to 1 each time a frame is received. 0: Frame not received 1: Frame received
17
RDE
0
R/W
Receive Descriptor Empty When receive descriptor empty (RDE = 1) occurs, receiving can be restarted by setting RACT = 1 in the receive descriptor and initiating receiving. 0: Receive descriptor active bit RACT = 1 not detected 1: Receive descriptor active bit RACT = 0 detected
16
RFOF
0
R/W
Receive FIFO Overflow Indicates that the receive FIFO has overflowed during frame reception. 0: Overflow has not occurred 1: Overflow has occurred
15 to 12
All 0
R
Reserved These bits are always read as 0. The write value should always be 0.
11
CND
0
R/W
Carrier Not Detect Indicates the carrier detection status. 0: A carrier is detected when transmission starts 1: A carrier is not detected when transmission starts
10
DLC
0
R/W
Detect Loss of Carrier Indicates that loss of the carrier has been detected during frame transmission. 0: Loss of carrier not detected 1: Loss of carrier detected
9
CD
0
R/W
Delayed Collision Detect Indicates that a delayed collision has been detected during frame transmission. 0: Delayed collision not detected 1: Delayed collision detected
Rev. 6.00 Jun. 12, 2007 Page 268 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
Bit 8
Bit Name TRO
Initial value 0
R/W R/W
Description Transmit Retry Over Indicates that a retry-over condition has occurred during frame transmission. Total 16 transmission retries including 15 retries based on the back-off algorithm are failed after the EtherC transmission starts. 0: Transmit retry-over condition not detected 1: Transmit retry-over condition detected
7
RMAF
0
R/W
Receive Multicast Address Frame 0: Multicast address frame has not been received 1: Multicast address frame has been received
6, 5
All 0
R
Reserved These bits are always read as 0. The write value should always be 0.
4
RRF
0
R/W
Receive Residual-Bit Frame 0: Residual-bit frame has not been received 1: Residual-bit frame has been received
3
RTLF
0
R/W
Receive Too-Long Frame Indicates that the frame more than the number of receive frame length upper limit set by RFLR of the EtherC has been received. 0: Too-long frame has not been received 1: Too-long frame has been received
2
RTSF
0
R/W
Receive Too-Short Frame Indicates that a frame of fewer than 64 bytes has been received. 0: Too-short frame has not been received 1: Too-short frame has been received
1
PRE
0
R/W
PHY Receive Error 0: PHY receive error not detected 1: PHY receive error detected
0
CERF
0
R/W
CRC Error on Received Frame 0: CRC error not detected 1: CRC error detected
Rev. 6.00 Jun. 12, 2007 Page 269 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
12.2.7
EtherC/E-DMAC Status Interrupt Permission Register (EESIPR)
EESIPR is a 32-bit readable/writable register that enables interrupts corresponding to individual bits in the EtherC/E-DMAC status register (EESR). An interrupt is enabled by writing 1 to the corresponding bit. In the initial state, interrupts are not enabled.
Bit 31 Bit Name Initial value 0 R/W R Description Reserved This bit is always read as 0. The write value should always be 0. 30 TWBIP 0 R/W Write-Back Complete Interrupt Permission 0: Write-back complete interrupt is disabled 1: Write-back complete interrupt is enabled 29 to 27 All 0 R Reserved These bits are always read as 0. The write value should always be 0. 26 TABTIP 0 R/W Transmit Abort Detection Interrupt Permission 0: Transmit abort detection interrupt is disabled 1: Transmit abort detection interrupt is enabled 25 RABTIP 0 R/W Receive Abort Detection Interrupt Permission 0: Receive abort detection interrupt is disabled 1: Receive abort detection interrupt is enabled 24 RFCOFIP 0 R/W Receive Frame Counter Overflow Interrupt Permission 0: Receive frame counter overflow interrupt is disabled 1: Receive frame counter overflow interrupt is enabled
Rev. 6.00 Jun. 12, 2007 Page 270 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
Bit 23
Bit Name ADEIP
Initial value 0
R/W R/W
Description Address Error Interrupt Permission 0: Address error interrupt is disabled 1: Address error interrupt is enabled
22
ECIIP
0
R/W
EtherC Status Register Interrupt Permission 0: EtherC status interrupt is disabled 1: EtherC status interrupt is enabled
21
TCIP
0
R/W
Frame Transmit Complete Interrupt Permission 0: Frame transmit complete interrupt is disabled 1: Frame transmit complete interrupt is enabled
20
TDEIP
0
R/W
Transmit Descriptor Empty Interrupt Permission 0: Transmit descriptor empty interrupt is disabled 1: Transmit descriptor empty interrupt is enabled
19
TFUFIP
0
R/W
Transmit FIFO Underflow Interrupt Permission 0: Underflow interrupt is disabled 1: Underflow interrupt is enabled
18
FRIP
0
R/W
Frame Received Interrupt Permission 0: Frame received interrupt is disabled 1: Frame received interrupt is enabled
17
RDEIP
0
R/W
Receive Descriptor Empty Interrupt Permission 0: Receive descriptor empty interrupt is disabled 1: Receive descriptor empty interrupt is enabled
16
RFOFIP
0
R/W
Receive FIFO Overflow Interrupt Permission 0: Receive FIFO overflow interrupt is disabled 1: Receive FIFO overflow interrupt is enabled
15 to 12
All 0
R
Reserved These bits are always read as 0. The write value should always be 0.
11
CNDIP
0
R/W
Carrier Not Detect Interrupt Permission 0: Carrier not detect interrupt is disabled 1: Carrier not detect interrupt is enabled
Rev. 6.00 Jun. 12, 2007 Page 271 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
Bit 10
Bit Name DLCIP
Initial value 0
R/W R/W
Description Detect Loss of Carrier Interrupt Permission 0: Detect loss of carrier interrupt is disabled 1: Detect loss of carrier interrupt is enabled
9
CDIP
0
R/W
Delayed Collision Detect Interrupt Permission 0: Delayed collision detect interrupt is disabled 1: Delayed collision detect interrupt is enabled
8
TROIP
0
R/W
Transmit Retry Over Interrupt Permission 0: Transmit retry over interrupt is disabled 1: Transmit retry over interrupt is enabled
7
RMAFIP
0
R/W
Receive Multicast Address Frame Interrupt Permission 0: Receive multicast address frame interrupt is disabled 1: Receive multicast address frame interrupt is enabled
6, 5
All 0
R
Reserved This bit is always read as 0. The write value should always be 0.
4
RRFIP
0
R/W
Receive Residual-Bit Frame Interrupt Permission 0: Receive residual-bit frame interrupt is disabled 1: Receive residual-bit frame interrupt is enabled
3
RTLFIP
0
R/W
Receive Too-Long Frame Interrupt Permission 0: Receive too-long frame interrupt is disabled 1: Receive too-long frame interrupt is enabled
2
RTSFIP
0
R/W
Receive Too-Short Frame Interrupt Permission 0: Receive too-short frame interrupt is disabled 1: Receive too-short frame interrupt is enabled
1
PREIP
0
R/W
PHY-LSI Receive Error Interrupt Permission 0: PHY-LSI receive error interrupt is disabled 1: PHY-LSI receive error interrupt is enabled
0
CERFIP
0
R/W
CRC Error on Received Frame 0: CRC error on received frame interrupt is disabled 1: CRC error on received frame interrupt is enabled
Rev. 6.00 Jun. 12, 2007 Page 272 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
12.2.8
Transmit/Receive Status Copy Enable Register (TRSCER)
TRSCER specifies whether or not transmit and receive status information reported by bits in the EtherC/E-DMAC status register is to be indicated in bits TFS26 to TFS0 and RFS26 to RFS0 in the corresponding descriptor. Bits in this register correspond to bits 11 to 0 in the EtherC/EDMAC status register (EESR). When a bit is cleared to 0, the transmit status (bits 11 to 8 in EESR) is indicated in bits TFS3 to TFS0 in the transmit descriptor, and the receive status (bits 7 to 0 in EESR) is indicated in bits RFS7 to RFS0 of the receive descriptor. When a bit is set to 1, the occurrence of the corresponding interrupt is not indicated in the descriptor. After this LSI is reset, all bits are cleared to 0.
Bit 31 to 12 Bit Name Initial value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 11 CNDCE 0 R/W CND Bit Copy Directive 0: Indicates the CND bit state in bit TFS3 in the transmit descriptor 1: Occurrence of the corresponding interrupt is not indicated in bit TFS3 of the transmit descriptor 10 DLCCE 0 R/W DLC Bit Copy Directive 0: Indicates the DLC bit state in bit TFS2 of the transmit descriptor 1: Occurrence of the corresponding interrupt is not indicated in bit TFS2 of the transmit descriptor 9 CDCE 0 R/W CD Bit Copy Directive 0: Indicates the CD bit state in bit TFS1 of the transmit descriptor 1: Occurrence of the corresponding interrupt is not indicated in bit TFS1 of the transmit descriptor 8 TROCE 0 R/W TRO Bit Copy Directive 0: Indicates the TRO bit state in bit TFS0 of the receive descriptor 1: Occurrence of the corresponding interrupt is not indicated in bit TFS0 of the receive descriptor
Rev. 6.00 Jun. 12, 2007 Page 273 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
Bit 7
Bit Name RMAFCE
Initial value 0
R/W R/W
Description RMAF Bit Copy Directive 0: Indicates the RMAF bit state in bit RFS7 of the receive descriptor 1: Occurrence of the corresponding interrupt is not indicated in bit RFS7 of the receive descriptor
6, 5
All 0
R
Reserved These bits are always read as 0. The write value should always be 0.
4
RRFCE
0
R/W
RRF Bit Copy Directive 0: Indicates the RRF bit state in bit RFS4 of the receive descriptor 1: Occurrence of the corresponding interrupt is not indicated in bit RFS4 of the receive descriptor
3
RTLFCE
0
R/W
RTLF Bit Copy Directive 0: Indicates the RTLF bit state in bit RFS3 of the receive descriptor 1: Occurrence of the corresponding interrupt is not indicated in bit RFS3 of the receive descriptor
2
RTSFCE
0
R/W
RTSF Bit Copy Directive 0: Indicates the RTSF bit state in bit RFS2 of the receive descriptor 1: Occurrence of the corresponding interrupt is not indicated in bit RFS2 of the receive descriptor
1
PRECE
0
R/W
PRE Bit Copy Directive 0: Indicates the PRF bit state in bit RFS1 of the receive descriptor 1: Occurrence of the corresponding interrupt is not indicated in bit RFS1 of the receive descriptor
0
CERFCE
0
R/W
CERF Bit Copy Directive 0: Indicates the CERF bit state in bit RFS0 of the receive descriptor 1: Occurrence of the corresponding interrupt is not indicated in bit RFS0 of the receive descriptor
Rev. 6.00 Jun. 12, 2007 Page 274 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
12.2.9
Receive Missed-Frame Counter Register (RMFCR)
RMFCR is a 16-bit counter that indicates the number of frames missed (discarded, and not transferred to the receive buffer) during reception. When the receive FIFO overflows, the receive frames in the FIFO are discarded. The number of frames discarded at this time is counted. When the value in this register reaches H'FFFF, counting-up is halted. When this register is read, the counter value is cleared to 0. Write operations to this register have no effect.
Bit 31 to 16 Bit Name Initial value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 15 to 0 MFC15 to MFC0 All 0 R Missed-Frame Counter Indicate the number of frames that are discarded and not transferred to the receive buffer during reception.
12.2.10 Transmit FIFO Threshold Register (TFTR) TFTR is a 32-bit readable/writable register that specifies the transmit FIFO threshold at which the first transmission is started. The actual threshold is 4 times the set value. The EtherC starts transmission when the amount of data in the transmit FIFO exceeds the number of bytes specified by this register, when the transmit FIFO is full, or when 1-frame write is executed. When setting this register, do so in the transmission-halt state.
Bit 31 to 11 Bit Name Initial value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0.
Rev. 6.00 Jun. 12, 2007 Page 275 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
Bit 10 to 0
Bit Name TFT10 to TFT0
Initial value All 0
R/W R/W
Description Transmit FIFO Threshold When setting a transmit FIFO, the FIFO must be set to a smaller value than the specified value of the FIFO capacity by FDR. H'00: Store and forward modes H'01 to H'0C: Setting prohibited H'0D: 52 bytes H'0E: 56 bytes : : H'1F: 124 bytes H'20: 128 bytes : : H'3F: 252 bytes H'40: 256 bytes H'41: 260 bytes (Setting prohibited in SH7618, setting enabled in SH7618A) H'42: 264 bytes (Setting prohibited in SH7618, setting enabled in SH7618A) : : H'7F: 508 bytes (Setting prohibited in SH7618, setting enabled in SH7618A) H'80: 512 bytes (Setting prohibited in SH7618, setting enabled in SH7618A) H'81 to H'200: Setting prohibited
Note: When starting transmission before one frame of data write has completed, take care the generation of the underflow.
Rev. 6.00 Jun. 12, 2007 Page 276 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
12.2.11 FIFO Depth Register (FDR) FDR is a 32-bit readable/writable register that specifies the capacity of the transmit and receive FIFOs.
Bit 31 to 11 Bit Name Initial value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 10 to 8 TFD2 to TFD0 B'000 R Transmit FIFO Capacity Specify the capacity of transmit FIFO. The set value should not be changed after the transmit/receive operation is started. 000: 256 bytes 001: 512 bytes (Setting prohibited in SH7618, setting enabled in SH7618A) Other than above: Setting prohibited 7 to 3 All 0 R Reserved These bits are always read as 0. The write value should always be 0. 2 to 0 RFD2 to RFD0 B'000 R Receive FIFO Capacity Specify the capacity of receive FIFO. The set value should not be changed after the transmit/receive operation is started. 000: 256 bytes 001: 512 bytes (Setting prohibited in SH7618, setting enabled in SH7618A) Other than above: Setting prohibited
Rev. 6.00 Jun. 12, 2007 Page 277 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
12.2.12 Receiving method Control Register (RMCR) RMCR is a 32-bit readable/writable register that specifies the control method for the RR bit in EDRRR when a frame is received. This register must be set during the receiving-halt state.
Bit 31 to 1 Bit Name Initial value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 0 RNC 0 R/W Receive Enable Control 0: When reception of one frame is completed, the EDMAC writes the receive status into the descriptor and clears the RR bit in EDRRR 1: When reception of one frame is completed, the EDMAC writes the receive status into the descriptor, reads the next descriptor, and prepares to receive the next frame
Rev. 6.00 Jun. 12, 2007 Page 278 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
12.2.13 E-DMAC Operation Control Register (EDOCR) EDOCR is a 32-bit readable/writable register that specifies the control methods used in E-DMAC operation.
Bit 31 to 4 Bit Name Initial value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 3 FEC 0 R/W FIFO Error Control Specifies E-DMAC operation when transmit FIFO underflow or receive FIFO overflow occurs. 0: E-DMAC operation continues when underflow or overflow occurs 1: E-DMAC operation halts when underflow or overflow occurs 2 AEC 0 R/W Address Error Control Indicates detection of an illegal memory address in an attempted E-DMAC transfer. 0: Illegal memory address not detected (normal operation) 1: E-DMAC stops its operation due to illegal memory address detection Note: To resume the operation, set the E-DMAC again after software reset by means of the SWR bit in EDMR. 1 EDH 0 R/W E-DMAC Halted 0: The E-DMAC is operating normally 1: The E-DMAC has been halted by NMI pin assertion. E-DMAC operation is restarted by writing 0 0 0 R Reserved This bit is always read as 0. The write value should always be 0.
Rev. 6.00 Jun. 12, 2007 Page 279 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
12.2.14 Receiving-Buffer Write Address Register (RBWAR) RBWAR stores the address of data to be written in the receiving buffer when the E-DMAC writes data to the receiving buffer. Which addresses in the receiving buffer are processed by the EDMAC can be recognized by monitoring addresses displayed in this register. The address that the E-DMAC is actually processing may be different from the value read from this register.
Bit 31 to 0 Bit Name RBWA31 to RBWA0 Initial value All 0 R/W R Description Receiving-Buffer Write Address These bits can only be read. Writing is prohibited.
12.2.15 Receiving-Descriptor Fetch Address Register (RDFAR) RDFAR stores the descriptor start address that is required when the E-DMAC fetches descriptor information from the receiving descriptor. Which receiving descriptor information is used for processing by the E-DMAC can be recognized by monitoring addresses displayed in this register. The address from which the E-DMAC is actually fetching a descriptor may be different from the value read from this register.
Bit 31 to 0 Bit Name RDFA31 to RDFA0 Initial value All 0 R/W R Description Receiving-Descriptor Fetch Address These bits can only be read. Writing is prohibited.
12.2.16 Transmission-Buffer Read Address Register (TBRAR) TBRAR stores the address of the transmission buffer when the E-DMAC reads data from the transmission buffer. Which addresses in the transmission buffer are processed by the E-DMAC can be recognized by monitoring addresses displayed in this register. The address from which the E-DMAC is actually reading in the buffer may be different from the value read from this register.
Bit 31 to 0 Bit Name TBRA31 to TBRA0 Initial value All 0 R/W R Description Transmission-Buffer Read Address These bits can only be read. Writing is prohibited.
Rev. 6.00 Jun. 12, 2007 Page 280 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
12.2.17 Transmission-Descriptor Fetch Address Register (TDFAR) TDFAR stores the descriptor start address that is required when the E-DMAC fetches descriptor information from the transmission descriptor. Which transmission descriptor information is used for processing by the E-DMAC can be recognized by monitoring addresses displayed in this register. The address from which the E-DMAC is actually fetching a descriptor may be different from the value read from this register.
Bit 31 to 0 Bit Name TDFA31 to TDFA0 Initial value All 0 R/W R Description Transmission-Descriptor Fetch Address These bits can only be read. Writing is prohibited.
12.2.18 Flow Control FIFO Threshold Register (FCFTR) FCFTR is a 32-bit readable/writable register that sets the flow control of the EtherC (setting the threshold on automatic PAUSE transmission). The threshold can be specified by the depth of the receive FIFO data (RFD2 to RFD0) and the number of receive frames (RFF2 to RFF0). The condition to start the flow control is decided by taking OR operation on the two thresholds. Therefore, the flow control by the two thresholds is independently started. When flow control is performed according to the RFD bits setting, if the setting is the same as the depth of the receive FIFO specified by the FIFO depth register (FDR), flow control is started when the remaining FIFO is (FIFO data depth - 64) bytes. For instance, when RFD in FDR = 0 and RFD in FCFTR = 0, flow control is started when (256 - 64) bytes of data is stored in the receive FIFO. The value set in the RFD bits in this register should be equal to or less than those in FDR.
Bit 31 to 19 Bit Name Initial value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0.
Rev. 6.00 Jun. 12, 2007 Page 281 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
Bit 18 17 16
Bit Name RFF2 RFF1 RFF0
Initial value 1 1 1
R/W R/W R/W R/W
Description Receive Frame Number Flow Control Threshold 000: When one receive frame has been stored in the receive FIFO 001: When two receive frames have been stored in the receive FIFO : : 110: When seven receive frames have been stored in the receive FIFO 111: When eight receive frames have been stored in the receive FIFO
15 to 3
All 0
Reserved These bits are always read as 0. The write value should always be 0.
2 1 0
RFD2 RFD1 RFD0
0 0 0
R R R
Receive Byte Flow Control Threshold 000: When (256 - 64) bytes of data is stored in the receive FIFO 001: When (512 - 64) bytes of data is stored in the receive FIFO (Setting prohibited in SH7618, setting enabled in SH7618A) Other than above: Setting prohibited
12.2.19 Transmit Interrupt Register (TRIMD) TRIMD is a 32-bit readable/writable register that specifies whether or not to notify write-back completion for each frame using the TWB bit in EESR and an interrupt on transmit operations.
Bit 31 to 1 Bit Name Initial value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 0 TIS 0 R/W Transmit Interrupt Setting 0: Write-back completion for each frame is not notified 1: Write-backed completion for each frame using the TWB bit in EESR is notified
Rev. 6.00 Jun. 12, 2007 Page 282 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
12.3
Operation
The E-DMAC is connected to the EtherC, and performs efficient transfer of transmit/receive data between the EtherC and memory (buffers) without the intervention of the CPU. The E-DMAC itself reads control information, including buffer pointers called descriptors, relating to the buffers. The E-DMAC reads transmit data from the transmit buffer and writes receive data to the receive buffer in accordance with this control information. By setting up a number of consecutive descriptors (a descriptor list), it is possible to execute transmission and reception continuously. 12.3.1 Descriptor List and Data Buffers
Before starting transmission/reception, the communication program creates transmit and receive descriptor lists in memory. The start addresses of these lists are then set in the transmit and receive descriptor list start address registers. The descriptor start address must be aligned so that it matches the address boundary according to the descriptor length set by the E-DMAC mode register (EDMR). The transmit buffer start address can be aligned with a byte, a word, and a longword boundary. (1) Transmit Descriptor
Figure 12.2 shows the relationship between a transmit descriptor and the transmit buffer. According to the specification in this descriptor, the relationship between the transmit frame and transmit buffer can be defined as one frame/one buffer or one frame/multi-buffer.
Transmit descriptor 31 30 29 28 27 26 TTTTT ADFFF CLPPE TE10 31 TD1 31 TD2 TBA Padding (4 bytes) TDL 0 0 TFS26 to TFS0 Valid transmit data Transmit buffer
TD0
16
Figure 12.2 Relationship between Transmit Descriptor and Transmit Buffer
Rev. 6.00 Jun. 12, 2007 Page 283 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
(a)
Transmit Descriptor 0 (TD0)
TD0 indicates the transmit frame status. The CPU and E-DMAC use RD0 to report the frame transmission status.
Bit 31 Bit Name TACT Initial value 0 R/W R/W Description Transmit Descriptor Active Indicates that this descriptor is active. The CPU sets this bit after transmit data has been transferred to the transmit buffer. The E-DMAC resets this bit on completion of a frame transfer or when transmission is suspended. 0: The transmit descriptor is invalid. Indicates that valid data has not been written to this bit by the CPU, or this bit has been reset by a write-back operation on termination of E-DMAC frame transfer processing (completion or suspension of transmission) If this state is recognized in an E-DMAC descriptor read, the E-DMAC terminates transmit processing and transmit operations cannot be continued (a restart is necessary) 1: The transmit descriptor is valid. Indicates that valid data has been written to the transmit buffer by the CPU and frame transfer processing has not yet been executed, or that frame transfer is in progress When this state is recognized in an E-DMAC descriptor read, the E-DMAC continues with the transmit operation 30 TDLE 0 R/W Transmit Descriptor List End After completion of the corresponding buffer transfer, the E-DMAC references the first descriptor. This specification is used to set a ring configuration for the transmit descriptors. 0: This is not the last transmit descriptor list 1: This is the last transmit descriptor list
Rev. 6.00 Jun. 12, 2007 Page 284 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
Bit 29 28
Bit Name TFP1 TFP0
Initial value 0 0
R/W R/W R/W
Description Transmit Frame Position 1, 0 These two bits specify the relationship between the transmit buffer and transmit frame. In the preceding and following descriptors, a logically positive relationship must be maintained between the settings of this bit and the TDLE bit. 00: Frame transmission for transmit buffer indicated by this descriptor continues (frame is not concluded) 01: Transmit buffer indicated by this descriptor contains end of frame (frame is concluded) 10: Transmit buffer indicated by this descriptor is start of frame (frame is not concluded) 11: Contents of transmit buffer indicated by this descriptor are equivalent to one frame (one frame/one buffer)
27
TFE
0
R/W
Transmit Frame Error Indicates that one or other bit of the transmit frame status indicated by bits 26 to 0 is set. Whether or not the transmit frame status information is copied into this bit is specified by the transmit/receive status copy enable register. 0: No error during transmission 1: An error occurred during transmission
26 to 0
TFS26 to TFS0
All 0
R/W
Transmit Frame Status TFS26 to TFS4: Reserved (The write value should always be 0.) TFS3: Carrier Not Detect (corresponds to CND bit in EESR) TFS2: Detect Loss of Carrier (corresponds to DLC bit in EESR) TFS1: Delayed collision Detect (corresponds to CD bit in EESR) TFS0: Transmit Retry Over (corresponds to TRO bit in EESR)
Rev. 6.00 Jun. 12, 2007 Page 285 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
(b)
Transmit Descriptor 1 (TD1)
TD1 specifies the transmit buffer length (maximum 64 kbytes).
Bit 31 to 16 Bit Name TDL Initial value All 0 R/W R/W Description Transmit Buffer Data Length These bits specify the valid transfer byte length in the corresponding transmit buffer. When the one frame/multi-buffer system is specified (TD0 and TFP = 10 or 00), the transfer byte length specified in the descriptors at the start and midway can be set in byte units. 15 to 0 All 0 R Reserved These bits are always read as 0. The write value should always be 0.
(c)
Transmit Descriptor 2 (TD2)
TD2 specifies the 32-bit transmit buffer start address. The transmit buffer start address setting can be aligned with a byte, a word, or a longword boundary.
Rev. 6.00 Jun. 12, 2007 Page 286 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
(2)
Receive Descriptor
Figure 12.3 shows the relationship between a receive descriptor and the receive buffer. In frame reception, the E-DMAC performs data rewriting up to a receive buffer 16-byte boundary, regardless of the receive frame length. Finally, the actual receive frame length is reported in the lower 16 bits of RD1 in the descriptor. Data transfer to the receive buffer is performed automatically by the E-DMAC to give a one frame/one buffer or one frame/multi-buffer configuration according to the size of one received frame.
Receive descriptor 31 30 29 28 27 26 RRRRR ADF FF CLPPE TE1 0 RBL 31 31 16 RBA Padding (4 bytes) 0 RFS26 to RFS0 Valid receive data Receive buffer
RD0
15 RDL
0 0
RD1 RD2
Figure 12.3 Relationship between Receive Descriptor and Receive Buffer
Rev. 6.00 Jun. 12, 2007 Page 287 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
(a)
Receive Descriptor 0 (RD0)
RD0 indicates the receive frame status. The CPU and E-DMAC use RD0 to report the frame receive status.
Bit 31 Bit Name RACT Initial value 0 R/W R/W Description Receive Descriptor Active Indicates that this descriptor is active. The E-DMAC resets this bit after receive data has been transferred to the receive buffer. On completion of receive frame processing, the CPU sets this bit to prepare for reception. 0: The receive descriptor is invalid. Indicates that the receive buffer is not ready (access disabled by E-DMAC), or this bit has been reset by a write-back operation on termination of E-DMAC frame transfer processing (completion or suspension of reception). If this state is recognized in an E-DMAC descriptor read, the E-DMAC terminates receive processing and receive operations cannot be continued. Reception can be restarted by setting RACT to 1 and executing receive initiation. 1: The receive descriptor is valid Indicates that the receive buffer is ready (access enabled) and processing for frame transfer from the FIFO has not been executed, or that frame transfer is in progress. When this state is recognized in an E-DMAC descriptor read, the E-DMAC continues with the receive operation. 30 RDLE 0 R/W Receive Descriptor List Last After completion of the corresponding buffer transfer, the E-DMAC references the first receive descriptor. This specification is used to set a ring configuration for the receive descriptors. 0: This is not the last receive descriptor list 1: This is the last receive descriptor list
Rev. 6.00 Jun. 12, 2007 Page 288 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
Bit 29 28
Bit Name RFP1 RFP0
Initial value 0 0
R/W R/W R/W
Description Receive Frame Position These two bits specify the relationship between the receive buffer and receive frame. 00: Frame reception for receive buffer indicated by this descriptor continues (frame is not concluded) 01: Receive buffer indicated by this descriptor contains end of frame (frame is concluded) 10: Receive buffer indicated by this descriptor is start of frame (frame is not concluded) 11: Contents of receive buffer indicated by this descriptor are equivalent to one frame (one frame/one buffer)
27
RFE
0
R/W
Receive Frame Error Indicates that one or other bit of the receive frame status indicated by bits 26 to 0 is set. Whether or not the receive frame status information is copied into this bit is specified by the transmit/receive status copy enable register. 0: No error during reception 1: A certain kind of error occurred during reception
Rev. 6.00 Jun. 12, 2007 Page 289 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
Bit 26 to 0
Bit Name RFS26 to RFS0
Initial value All 0
R/W R/W
Description Receive Frame Status These bits indicate the error status during frame reception. RFS26 to RFS10: Reserved (The write value should always be 0.) RFS9: Receive FIFO overflow (corresponds to RFOF bit in EESR) RFS8: Reserved (The write value should always be 0.) RFS7: Multicast address frame received (corresponds to RMAF bit in EESR) RFS6: CAM entry unregistered frame received (corresponds to the RUAF bit in EESR) RSF5: Reserved (The write value should always be 0.) RFS4: Receive residual-bit frame error (corresponds to RRF bit in EESR) RFS3: Receive too-long frame error (corresponds to RTLF bit in EESR) RFS2: Receive too-short frame error (corresponds to RTSF bit in EESR) RFS1: PHY-LSI receive error (corresponds to PRE bit in EESR) RFS0: CRC error on received frame (corresponds to CERF bit in EESR)
Rev. 6.00 Jun. 12, 2007 Page 290 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
(b)
Receive Descriptor 1 (RD1)
RD1 specifies the receive buffer length (maximum 64 kbytes).
Bit 31 to 16 Bit Name RBL Initial value All 0 R/W R/W Description Receive Buffer Length These bits specify the maximum reception byte length in the corresponding receive buffer. The transfer byte length must align with a 16-byte boundary (bits 19 to 16 cleared to 0). The maximum receive frame length with one frame per buffer is 1,514 bytes, excluding the CRC data. Therefore, for the receive buffer length specification, a value of 1,520 bytes (H'05F0) that takes account of a 16-byte boundary is set as the maximum receive frame length. 15 to 0 RDL All 0 R/W Receive Data Length These bits specify the data length of a receive frame stored in the receive buffer. The receive data transferred to the receive buffer does not include the 4-byte CRC data at the end of the frame. The receive frame length is reported as the number of words (valid data bytes) not including this CRC data.
(c)
Receive Descriptor 2 (RD2)
RD2 specifies the 32-bit receive buffer start address. The receive buffer start address must be aligned with a longword boundary. However, when SDRAM is connected, it must be aligned with a 16-byte boundary. 12.3.2 Transmission
When the transmit function is enabled and the transmit request bit (TR) is set in the E-DMAC transmit request register (EDTRR), the E-DMAC reads the descriptor used last time from the transmit descriptor list (in the initial state, the descriptor indicated by the transmission descriptor start address register (TDLAR)). If the setting of the TACT bit in the read descriptor is active, the E-DMAC reads transmit frame data sequentially from the transmit buffer start address specified by TD2, and transfers it to the EtherC. The EtherC creates a transmit frame and starts transmission to the MII. After DMA transfer of data equivalent to the buffer length specified in the descriptor, the following processing is carried out according to the TFP value.
Rev. 6.00 Jun. 12, 2007 Page 291 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
1. TFP = 00 or 01 (frame continuation): Descriptor write-back is performed after DMA transfer. 2. TFP = 01 or 11 (frame end): Descriptor write-back is performed after completion of frame transmission. The E-DMAC continues reading descriptors and transmitting frames as long as the setting of the TACT bit in the read descriptors is "active." When a descriptor with an "inactive" TACT bit is read, the E-DMAC resets the transmit request bit (TR) in the transmit register and ends transmit processing (EDTRR).
Transmission flowchart This LSI + memory E-DMAC Transmit FIFO EtherC PHY
EtherC/E-DMAC initialization
Descriptor and transmit buffer setting Transmit directive Descriptor read
Transmit data transfer Descriptor write-back Descriptor read
Transmit data transfer Frame transmission
Descriptor write-back Transmission completed
Figure 12.4 Sample Transmission Flowchart
Rev. 6.00 Jun. 12, 2007 Page 292 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
12.3.3
Reception
When the receive function is enabled and the CPU sets the receive request bit (RR) in the EDMAC receive request register (EDRRR), the E-DMAC reads the descriptor following the previously used one from the receive descriptor list (in the initial state, the descriptor indicated by the transmission descriptor start address register (TDLAR)), and then enters the receive-standby state. If the setting of the RACT bit is "active" and an own-address frame is received, the EDMAC transfers the frame to the receive buffer specified by RD2. If the data length of the received frame is greater than the buffer length given by RD1, the E-DMAC performs write-back to the descriptor when the buffer is full (RFP = 10 or 00), then reads the next descriptor. The EDMAC then continues to transfer data to the receive buffer specified by the new RD2. When frame reception is completed, or if frame reception is suspended because of a certain kind of error, the E-DMAC performs write-back to the relevant descriptor (RFP = 11 or 01), and then ends the receive processing. The E-DMAC then reads the next descriptor and enters the receive-standby state again. To receive frames continuously, the receive enable control bit (RNC) must be set to 1 in the receive control register (RCR). After initialization, this bit is cleared to 0.
Rev. 6.00 Jun. 12, 2007 Page 293 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
Reception flowchart This LSI + memory E-DMAC Receive FIFO EtherC PHY
EtherC/E-DMAC initialization
Descriptor and receive buffer setting Start of reception Descriptor read
Frame reception
Receive data transfer Descriptor write-back Descriptor read
Receive data transfer Descriptor write-back Descriptor read (receive ready for the next frame)
Reception completed
Figure 12.5 Sample Reception Flowchart
Rev. 6.00 Jun. 12, 2007 Page 294 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
12.3.4
Multi-Buffer Frame Transmit/Receive Processing
Multi-Buffer Frame Transmit Processing If an error occurs during multi-buffer frame transmission, the processing shown in figure 12.6 is carried out by the E-DMAC. Where the transmit descriptor is shown as inactive (TACT bit = 0) in the figure, buffer data has already been transmitted normally, and where the transmit descriptor is shown as active (TACT bit = 1), buffer data has not been transmitted. If a frame transmit error occurs in the first descriptor part where the transmit descriptor is active (TACT bit = 1), transmission is halted, and the TACT bit cleared to 0, immediately. The next descriptor is then read, and the position within the transmit frame is determined on the basis of bits TFP1 and TFP0 (continuing [B'00] or end [B'01]). In the case of a continuing descriptor, the TACT bit is cleared to 0, only, and the next descriptor is read immediately. If the descriptor is the final descriptor, not only is the TACT bit cleared to 0, but write-back is also performed to the TFE and TFS bits at the same time. Data in the buffer is not transmitted between the occurrence of an error and write-back to the final descriptor. If error interrupts are enabled in the EtherC/E-DMAC status interrupt permission register (EESIPR), an interrupt is generated immediately after the final descriptor write-back.
Descriptors T A C T T D L E T F P 1 T F P 0
00 00 00 Inactivates TACT (change 1 to 0) E-DMAC Descriptor read Inactivates TACT Descriptor read Inactivates TACT Descriptor read Inactivates TACT Descriptor read Inactivates TACT and writes TFE, TFS 10 10 10 10 10 11
10 00 00 00 00 00 00 01 10 Buffer
Untransmitted data is not transmitted after error occurrence Descriptor is only processed.
Transmit error occurrence
One frame
Transmitted data Untransmitted data
Figure 12.6 E-DMAC Operation after Transmit Error
Rev. 6.00 Jun. 12, 2007 Page 295 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
Multi-Buffer Frame Receive Processing If an error occurs during multi-buffer frame reception, the processing shown in figure 12.7 is carried out by the E-DMAC. Where the receive descriptor is shown as inactive (RACT bit = 0) in the figure, buffer data has already been received normally, and where the receive descriptor is shown as active (RACT bit = 1), this indicates a buffer for which reception has not yet been performed. If a frame receive error occurs in the first descriptor part where the RACT bit = 1 in the figure, reception is halted immediately and a status write-back to the descriptor is performed. If error interrupts are enabled in the EtherC/E-DMAC status interrupt permission register (EESIPR), an interrupt is generated immediately after the write-back. If there is a new frame receive request, reception is continued from the buffer after that in which the error occurred.
Descriptors R A C T 0 0 0 Inactivates RACT and writes RFE, RFS E-DMAC 1 Descriptor read Write-back 1 1 1 1 1 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 Buffer New frame reception continues from buffer 0 0 1 R D L E 0 0 0 R F P 1 1 0 0 R F P 0 0 0 0 Receive error occurrence Start of frame
Figure 12.7 E-DMAC Operation after Receive Error
Rev. 6.00 Jun. 12, 2007 Page 296 of 610 REJ09B0131-0600
.........
Received data Unreceived data
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
12.4
12.4.1
Usage Notes
Usage Notes on SH-Ether EtherC/E-DMAC Status Register (EESR)
When the status bits in EESR of the on-chip E-DMAC of the SH-Ether chip are used as interrupt sources, setting of the interrupt source may fail if software writes a 1 to the corresponding status bit in EESR to clear the bit and this coincides with setting of the status interrupt source in EESR by the EtherC or E-DMAC. Figure 12.8 shows an example of timing in the case where setting of the interrupt source in EESR has failed. (a) In this example, both the reception interrupt and transmission interrupt sources of EESR are used. Firstly, reception interrupt source A from the EtherC or E-DMAC sets bit A in EESR and an interrupt is generated. (b) The interrupt handler writes 1 to bit A to clear it. (c) If clearing of bit A by writing of a 1 and generation of the transmission-interrupt source B signal by the EtherC or E-DMAC take place simultaneously, bit A will be cleared but the status bit for transmission-interrupt source B in EESR might not be set.
(a) (b), (c) (c)
Internal clock (I) Reception interrupt source A generated by EtherC/E-DMAC Transmission interrupt source B generated by EtherC/E-DMAC
Simultaneous clearing of the bit by writing of a 1 and generation of interrupt source B.
Bit A in EESR Bit B in EESR Write access to EESR by software Data to be written to EESR
H'00000001
Reception interrupt source A is set in bit A of EESR.
Only bit A of EESR is cleared by software.
Failure to generate transmission interrupt source B due to non-setting of bit B.
Bit clearing by writing a 1 : Expected operation
Figure 12.8 Timing of the Case where Setting of the Interrupt Source Bit in EESR by the EDMAC Fails
Rev. 6.00 Jun. 12, 2007 Page 297 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
(1)
Countermeasure
This problem does not occur with all of the bits in EESR. The description applies to some bits but not others. Table 12.1 shows whether the problem can occur with the individual bits and whether the state of the individual interrupt source is reflected in the descriptor. Table 12.1 EESR Bits for which This Problem can Occur and Reflection of Interrupt Sources in the Descriptor
Bit 31 30 29 28 27 26 Bit Name Status TWB TABT Reserved Write-back complete Reserved Reserved Reserved Transmit abort detected Possibility of Problem Yes Yes Reflection in Descriptor Reflected in TD0 bit8 (TFS8) Reflected in RD0 bit8 (RFS8) Reflected in TD0 bit31 (TACT) Reflected in RD0 bit31 (RACT) Reflected in RD0 bit9 (RFS9) Interrupt Source Transmit Transmit
25
RABT
Receive abort detected
No
Reception
24 23 22 21
RFCOF ADE ECI TC
Receive frame counter overflow Address error EtherC status register interrupt source Frame transmission complete
Yes No No Yes
Reception Others Others Transmit
20 19 18
TDE TFUF FR
Transmit descriptor empty Transmit FIFO underflow Frame received
No Yes No
Transmit Transmit Reception
17 16
RDE RFOF
Receive descriptor empty Receive FIFO overflow
No Yes
Reception Reception
Rev. 6.00 Jun. 12, 2007 Page 298 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
Bit 15 14 13 12
Bit Name Status Reserved Reserved Reserved Transmit frame length error
Possibility of Problem Yes
Reflection in Descriptor Reflected in TD0 bit4 (TFS4) Reflected in TD0 bit3 (TFS3) Reflected in TD0 bit2 (TFS2) Reflected in TD0 bit1 (TFS1) Reflected in TD0 bit0 (TFS0) Reflected in RD0 bit7 (RFS7) Reflected in RD0 bit5 (RFS5) Reflected in RD0 bit4 (RFS4) Reflected in RD0 bit3 (RFS3) Reflected in RD0 bit2 (RFS2) Reflected in RD0 bit1 (RFS1)
Interrupt Source Transmit
11
CND
Carrier not detected
Yes
Transmit
10
DLC
Loss of carrier detected
Yes
Transmit
9
CD
Delayed collision detected
Yes
Transmit
8
TRO
Transmit retry over
Yes
Transmit
7
RMAF
Multicast address frame received
No
Reception
6 5

Reserved Receive frame discard request asserted Residual-bit frame received
No
Reception
4
RRF
No
Reception
3
RTLF
Overly long frame received
No
Reception
2
RTSF
Overly short frame received
No
Reception
1
PRE
PHY receive error
No
Reception
Rev. 6.00 Jun. 12, 2007 Page 299 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
Bit 0
Bit Name Status CERF CRC error in received frame
Possibility of Problem No
Reflection in Descriptor Reflected in RD0 bit0 (RFS0)
Interrupt Source Reception
"Yes": Setting of this interrupt source bit can fail. "No": Setting of this interrupt source bit does not fail.
Take the following countermeasures for bits where the problem can arise. * Bit 30 (TWB): Write-back complete interrupt source bit in EESR may not be set. Check the TACT bit in the transmit descriptor. TACT = 0 indicates that the transmission is complete. * Bit 26 (TABT): Transmit abort detection interrupt source bit in EESR may not be set. Since the state of the interrupt source is written back to the relevant descriptor, check the transmit descriptor (TD0) to confirm the error status. * Bit 24 (RFCOF): Receive frame counter overflow interrupt source bit in EESR may not be set. However, even if the software is not notified of the interrupt despite the frame counter having overflowed, the upper layer (e.g. TCP/IP) can recognize the error because this LSI discards the frame. After departure from the overflow state, storage in the receive FIFO proceeds normally from the head of the next frame. Therefore, no problem with the system arises. * Bit 21 (TC): Frame transmission complete interrupt source bit in EESR may not be set. For transmission-related processing, either procedure (a) or (b) given below is effective. (a) Transmission processing without interrupt handling of the frame transmission complete interrupt 1. Prepare multiple transmit descriptors so that multiple frames can be transmitted. 2. After setting the transmit descriptors, set bit 0 (TR) in the E-DMAC transmit request register (EDTRR) to start transmission. 3. Before setting the next frame for transmission in the descriptor (when a transmission task arises), check the TACT bit of the corresponding transmit descriptor. 4. If the TACT bit is clear, set the frame for transmission in the corresponding transmit descriptor and set the TR bit in EDTRR to start transmission. If the TACT bit is set to 1, do not set the transmit descriptor until the next timing. (b) For systems where completion of the transmission of each frame must be confirmed (that is, set frame for transmission initiate transmission complete frame transmission set the next frame for transmission ...) 1. Check the TACT bit in the last descriptor of the frame for transmission and confirm that TACT = 0, which means that the transmission was completed.
Rev. 6.00 Jun. 12, 2007 Page 300 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
* Bit 19 (TFUF): Transmit FIFO underflow interrupt source bit in EESR may not be set. When this bit is used as an interrupt source but is not set when it should be, the software is not notified of the interrupt. However, the upper layer will recognize the error in the form of an underflow of the transmit FIFO. * Bit 16 (RFOF): Receive FIFO overflow interrupt source bit in EESR may not be set. Since the state of the interrupt source is written back to the relevant descriptor, check the receive descriptor (RD0) to confirm the error status. * Bit 11 (CND), bit 10 (DLC), bit 9 (CD), bit 8 (TRO): The interrupt source bits in EESR for the carrier not detected, loss of carrier detected, delayed collision detected, and transmit retry over interrupts may not be set. However, since the states of the interrupt sources are written back to the relevant descriptor, check the transmit descriptor (TD0) to confirm the error status. (2) Example of a countermeasure when the software configuration is based on the frame transmit complete interrupt
The following descriptions are of sample countermeasures for cases when software processing is based on the frame transmit complete interrupt (bit 21 (TC) in EESR). If the TC interrupt source bit (bit 21) in EESR is not set on completion of transmission, the system will continue to wait for the TC interrupt, leading to stoppage of transmission. This situation arises when the interrupt handler writes a 1 to clear the bit. The sample method given as case (a) below takes the above possibility into account and avoids the problem by monitoring the transmit descriptor in interrupt processing for interrupts other than the TC interrupt. The sample method given as case (b) below avoids the above problem by setting a timeout limit for retry processing when multiple transmit descriptors are in use. Note: The countermeasure should be the one that best suits the structure of your driver and other software.
Rev. 6.00 Jun. 12, 2007 Page 301 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
(a)
Countermeasure by monitoring of the transmit descriptor in the processing of interrupts other than the frame transmit complete (TC) interrupt
1. Prepare multiple transmit descriptors so that multiple frames can be transmitted. 2. Provide a "condition flag" for use in step 5 and by interrupt handlers, and then turn off this flag. This flag serves as a condition flag into which the TACT bits of transmit descriptors are read out. 3. After setting the frame for transmission in the first descriptor, start transmission by setting bit 0 (TR) in the E-DMAC transmit request register (EDTRR). 4. Before setting the next frame for transmission in the transmit descriptor (when another transmission task arises), check the TACT bit in the corresponding transmit descriptor. 5. If the TACT bit is clear, set the frame for transmission in the corresponding transmit descriptor and start transmission by setting the TR bit in EDTRR. If the TACT bit is set to 1, turn on the condition flag and make an OS service call (e.g. to acquire the semaphore) to place the transmission task in the waiting state. Note: Before setting the TR bit in EDTRR, always read the TR bit and make sure that TR = 0. 6. Wait until the transmission task leaves the waiting state. There are two conditions for making the OS service call (e.g. returning the semaphore) from the interrupt handler to take the task out of the waiting state. Generation of a TC interrupt Generation of an interrupt other than the TC interrupt while the condition flag is on and TACT = 0. Elimination of unwanted processing by checking the TACT bit is only possible when the condition flag is on. The condition flag should be turned off after the task has left the waiting state. 7. When the transmission task has left the waiting state and entered execution, set the transmit frame in the corresponding transmit descriptor and then set the TR bit in EDTRR to start transmission.
Rev. 6.00 Jun. 12, 2007 Page 302 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
Transmission task
1. 2. Prepare multiple transmit descriptors. Prepare the condition flag and turn it off. Transmission starts After setting the transmit 3. descriptor, set the TR bit in EDTRR to 1.
Interrupt handler
Generation of EtherC/E-DMAC interrupt Save EESR and clear the bit by writing a 1. No
TC interrupt? Yes No End Interrupt other than TC? Yes Interrupt processing for interrupts other than TC No Is the condition flag on?
Next transmission task generated? Yes Read the TACT bit of the corresponding transmit descriptor.
Make an OS service call to bring the transmission task out of the waiting state.
No
4.
TACT = 0? Yes 5. Turn the condition flag on. Make an OS service call to place the transmission task in a waiting state.
No
Yes Read the TACT bit of the corresponding transmit descriptor. No TACT = 0?
5. After setting the corresponding transmit 7. descriptor, set the TR bit in EDTRR to 1.
6.
Has the transmission task been brought out of the waiting state by the interrupt handler? Yes
No
Yes Make an OS service call to bring the transmission task out of the waiting state. Turn off the condition flag.
TC: EESR frame transmission complete : Processing added as the countermeasure for the problem
End
Figure 12.9 Countermeasure by Monitoring the Transmit Descriptor in Processing of Interrupts Other than the Frame Transmit Complete (TC) Interrupt
Rev. 6.00 Jun. 12, 2007 Page 303 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
(b)
Countermeasure by adding timeout processing
1. Prepare multiple transmit descriptors so that multiple frames can be transmitted. 2. After setting the descriptors, set bit 0 (TR) in the E-DMAC transmit request register (EDTRR) to start transmission. 3. Before setting the next frame for transmission in the transmit descriptor (when a transmission task arises), check the TACT bit in the corresponding transmit descriptor. 4. If the TACT bit is clear, set the frame for transmission in the corresponding transmit descriptor and set the TR bit in EDTRR to start transmission. If the TACT bit is set to 1, place the transmission task in a waiting state by making an OS service call of a routine with a timeout function (e.g. acquire a semaphore that has a timeout). Note: Before setting the TR bit in EDTRR, always read the TR bit and make sure that TR = 0. 5. When the transmission task has left the waiting state and entered the execution state within the time limit, set the frame for transmission in the corresponding transmit descriptor and then set the TR bit in EDTRR to start transmission. Taking the transmission task out of the waiting state should be done by the interrupt handler when the TC interrupt is generated. 6. When the timeout limit is reached, check the TACT bit in the corresponding transmit descriptor. If the TACT bit is clear, set the frame for transmission in the corresponding transmit descriptor and set the TR bit in EDTRR to start transmission. If the TACT bit is set to 1, place the transmission task in a waiting state by making an OS service call of a routine with a timeout function, or execute a software reset to initialize all of the modules associated with Ethernet functionality.
Rev. 6.00 Jun. 12, 2007 Page 304 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
Transmission task
1.
Interrupt handler
Generation of EtherC/E-DMAC interrupt Save EESR and clear the bit by writing a 1.
No
Prepare multiple transmit descriptors. Transmission starts.
2.
After setting the transmit descriptor, set the TR bit in EDTRR to 1. TC interrupt? Next transmission task generated?
Yes No Yes
End
Make an OS service call to bring the transmission task out of the waiting state.
Read the TACT bit of the corresponding transmit descriptor.
Interrupt other than TC?
Yes
No
3.
No TACT = 0? Yes
Interrupt processing for interrupts other than TC
End
4. After setting the corresponding transmit 5. descriptor, set the TR bit in EDTRR to 1.
4. Place the transmission task in a waiting
state by calling an OS service routine with a timeout function.
Has the transmission task left the waiting state within the specified time?
Yes
No
Timeout
6.
Read the TACT bit of the corresponding transmit descriptor.
No TACT = 0? Yes
TC: EESR frame transmission complete : Processing added as the countermeasure for the problem
Figure 12.10 Method of Adding Timeout Processing
Rev. 6.00 Jun. 12, 2007 Page 305 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
12.4.2
Usage Notes on SH-Ether Transmit-FIFO Underflow
In the transmission operation of the on-chip E-DMAC of the SH-Ether, if the E-DMAC cannot acquire bus-mastership due to occupancy of the bus by a bus master other than the E-DMAC, data are not writable to the transmit FIFO and an underflow occurs. The expected operation from that point is as follows: on obtaining the bus mastership, the E-DMAC resumes transmission of the remaining data for transmission; on completion of the DMA transfer, it writes back to the corresponding descriptor, and then fetches the next transmit descriptor. However, if the size of the transmit FIFO set by the FIFO depth register (FDR) maximum frame length for transmission (1518 bytes), the E-DMAC may stop operating even if the transmit request bit (TR) in the EDMAC transmit request register (EDTRR) is set to 1, according to the relationship between the length of the remaining frame data and the value of the transmit FIFO pointer. The relationship between the stoppage of E-DMAC operation and the state of the transmit FIFO is shown below. The data for transmission, which are placed in external memory (transmit buffer), are DMAtransferred by the E-DMAC to the transmit FIFO and output from the MII pin via the EtherC module. The transmit FIFO write pointer (WP) is used when the E-DMAC writes the data for transmission to the transmit FIFO, and the transmit FIFO read pointer (RP) is used when the EtherC module reads the data for transmission from the transmit FIFO. 1. After a software reset, the transmit FIFO will have been initialized, and WP and RP will hold the minimum and maximum values, respectively, of the transmit FIFO capacity. 2. When the E-DMAC starts DMA transfer, WP is incremented when the data for transmission are written to the transmit FIFO. On the other hand, RP is incremented when the data written to the transmit FIFO are read out by the EtherC module. Note: The transmit FIFO only stores the data of a single frame that is being processed. It does not store data extending over multiple frames. This means that the E-DMAC does not transfer the next frame to the transmit FIFO until the data of the frame being processed are read from the transmit FIFO. 3. If the E-DMAC fails to get the bus mastership for a system-related reason, the DMA transfer does not proceed and a transmit underflow occurs (WP = RP < frame length). Read access to the transmit FIFO by the EtherC is then terminated and RP is initialized (to the maximum value of the size of the transmit FIFO). 4. On again acquiring the bus mastership, the E-DMAC resumes DMA transfer of the remaining data of the frame. However, if the transmit FIFO becomes full despite a failure to write all of the remaining frame data from the point when the transmit FIFO underflowed, the E-DMAC waits for the transmit FIFO to become empty before transferring further remaining data.
Rev. 6.00 Jun. 12, 2007 Page 306 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
However, as stated in step 3, the read access to the transmit FIFO by the EtherC module will have been terminated, and the E-DMAC thus stops operating with the transmit FIFO full. In short, this problem arises when [initial value of RP - WP value < length of remaining frame data] at the point of the transmit underflow.
Data for transmission is written by the E-DMAC. Transmit FIFO Maximum transmit FIFO capacity RP Transmit FIFO
WP Increment Minimum WP transmit FIFO capacity
Increment RP
Data for transmission is read by EtherC 1. Initial state after software reset 2. Writing and reading of the data for transmission
Data for transmission is written by the E-DMAC. Transmit FIFO Maximum transmit FIFO capacity WP RP Initial state RP WP Transmit FIFO Transmit FIFO is full RP
Minimum transmit FIFO capacity Data for transmission is read by EtherC Reading of data for transmission by EtherC is terminated. 4. Point where the problem makes the E-DMAC stop WP: Transmit FIFO write pointer RP: Transmit FIFO read pointer
3. Transmit-FIFO underflow has occurred
Figure 12.11 Operation when E-DMAC Stops and the Transmit FIFO
Rev. 6.00 Jun. 12, 2007 Page 307 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
(1)
Countermeasure
This problem occurs under this condition: size of transmit FIFO set in the FIFO depth register (FDR) maximum length of frame for transmission (1518bytes). To release the E-DMAC from the stopped state due to this problem, execute a software reset to initialize both the E-DMAC and EtherC modules. Specific countermeasures are given below. An example for the case where the software does not use TC interrupts in transmission processing is given as (2), and an example for the case with TC interrupt-driven software is given as (3). Both methods require the addition of timeout processing with a maximum specified time as the timeout limit, and are based on the countermeasures explained in section 12.4.1, Usage Notes on SH-Ether EtherC/E-DMAC Status Register (EESR). The constant specified time corresponds to the timeout limit stated in section 12.4.1, Usage Notes on SH-Ether EtherC/E-DMAC Status Register (EESR). The maximum specified time should be set with reference to the maximum times taking retry processing into consideration, as given in table 12.2. Derive n, the number of repetitions of the constant specified time, from this maximum specified time. If transfer takes more than the maximum specified time, this indicates that the EDMAC has stopped due to a transmission underflow. In this case, execute a software reset to initialize the EtherC and E-DMAC modules. Since the receiving side will also be initialized by the software reset, the receiving side may require processing in a higher-level layer (e.g. TCP/IP). Note: The countermeasure should be the one that best suits the structure of your driver and other software. (2) Countermeasure for the case where the software handles transmission without the aid of TC interrupts
The countermeasure described under (a), Processing transmission without handling of the frame transmission complete (TC) interrupt, below, is based on the method explained in the description of bit 21 in (1) of section 12.4.1, Usage Notes on SH-Ether EtherC/E-DMAC Status Register (EESR).
Rev. 6.00 Jun. 12, 2007 Page 308 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
(a) Processing transmission without handling of the frame transmission complete (TC) interrupt 1. Make initial settings for the timer. 2. Prepare multiple transmit descriptors so that multiple frames can be transmitted. 3. After setting the transmit descriptors, start transmission by setting bit 0 (TR) in the E-DMAC transmit request register (EDTRR). 4. Before setting the next frame for transmission in the transmit descriptor (when a transmission task arises), check the TACT bit in the corresponding transmit descriptor. 5. If the TACT bit is clear, set the frame for transmission in the corresponding transmit descriptor and start transmission by setting the TR bit in EDTRR. If the TACT bit is set to 1, set counter i to 0 (counter i is the variable that indicates the number of repetitions of the timer operation to measure the specified constant period). 6. Start counting by the timer. 7. When the specified constant period has elapsed, stop the timer counter and check the TACT bit in the corresponding transmit descriptor. 8. If the TACT bit is clear, set the frame for transmission in the corresponding transmit descriptor and set the TR bit in EDTRR to start transmission. If the TACT bit is set to 1, increment counter i. 9. While the TACT bit is found to be 1 in step 8 and the value of counter i is less than n, repeat steps 6 to 8 until the maximum specified time is reached (the maximum specified time should be set with reference to the maximum times in consideration of retry processing given in table 12.2, and from this maximum specified time, determine n, the number of repetitions of the specified constant period; n is determined by the user with reference to table 12.2). If counter i reaches or exceeds n, the maximum specified time has elapsed and we can judge that the E-DMAC has stopped due to a transmit underflow. Initialize the EtherC and E-DMAC modules by setting the software-reset bit SWR in the E-DMAC mode register (EDMR). After re-making initial settings for the Ethernet module, initialize the transmit/receive descriptors and transmit/receive buffers.
Rev. 6.00 Jun. 12, 2007 Page 309 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
Transmission task
1. 2.
Make initial settings for the timer. Prepare multiple transmit descriptors.
Transmission starts.
3.
After setting the transmit descriptor, set the TR bit in EDTRR to 1.
Next transmission task generated?
Yes
No
End
Read the TACT bit of the corresponding transmit descriptor.
4.
No TACT = 0? Yes 5.
i = 0;
5. After setting the corresponding transmit 8. descriptor, set the TR bit in EDTRR to 1.
6.
Start the timer.
Specified constant 1 period elapsed? *
Yes 7.
No
Stop the timer. Read the TACT bit of the corresponding transmit descriptor.
No TACT = 0? Yes i++;
8.
9.
i >= n? * Yes
2
No
Issue a software reset to initialize the EtherC and E-DMAC modules. Make initial settings of the EtherC and E-DMAC modules. Initialize the transmit/receive descriptors and transmit/receive buffers.
Notes: 1. The specified constant period is the timeout period mentioned in section 12.4.1, Usage Notes on SH-Ether EtherC/E-DMAC Status Register(EESR). 2. Set n with reference to the maximum specified time values in table 12.2. : Processing added as the countermeasure for the problem
Figure 12.12 Processing Transmission without Handling of the TC Interrupt
Rev. 6.00 Jun. 12, 2007 Page 310 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
(3)
Countermeasure for the case of TC interrupt-driven software
The sample countermeasure for the case of TC interrupt-driven software shown below is the addition of timeout processing within the limit imposed by the maximum specified time. This is based on the method explained in (b) Countermeasure by adding timeout processing in section 12.4.1, Usage Notes on SH-Ether EtherC/E-DMAC Status Register (EESR). The maximum specified time should be set with reference to the maximum times in consideration of retry processing (table 12.2). From this maximum specified time, determine n, the number of calls of the OS service routine with a timeout function. (b) Countermeasure as the addition of timeout processing within the limit imposed by the maximum specified time 1. Prepare multiple transmit descriptors so that multiple frames can be transmitted. 2. After setting the transmit descriptors, start transmission by setting bit 0 (TR) in the E-DMAC transmit request register (EDTRR). 3. Before setting the next frame for transmission in the transmit descriptor (when a transmission task arises), check the TACT bit in the transmit descriptor. 4. If the TACT bit is clear, set the frame for transmission in the corresponding transmit descriptor and start transmission by setting the TR bit in EDTRR. If the TACT bit is set to 1, set counter i to 0 (counter i is the variable that indicates the number of calls of the OS service routine with a timeout function). Then, place the transmission task in a waiting state by calling the OS routine (e.g. acquire a semaphore that has a timeout limit). Note: Before setting the TR bit in EDTRR, always read the TR bit and make sure that TR = 0. 5. When the transmission task has left the waiting state and entered the execution state within the specified constant period, set the frame for transmission in the corresponding transmit descriptor and then set the TR bit in EDTRR to start transmission. The transmission task should be taken out of the waiting state by the interrupt handler initiated by generation of the TC interrupt. 6. If the transmission task has not left the waiting state within the specified constant period, increment counter i. Then, if i < n, check the TACT bit in the corresponding transmit descriptor. The value for counting, n, is determined by the user with reference to table 12.2. 7. If the TACT bit is clear, set the frame for transmission in the corresponding transmit descriptor and set the TR bit in EDTRR to start transmission. If the TACT bit is set to 1, return the transmission task to the waiting state by calling an OS service routine that has a timeout function, and then repeat steps 5 and 6.
Rev. 6.00 Jun. 12, 2007 Page 311 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
8. If counter i reaches or exceeds n, the maximum specified time has elapsed and we can judge that the E-DMAC has stopped due to a transmit underflow. Initialize the EtherC and E-DMAC modules by setting the software-reset bit SWR in the E-DMAC mode register (EDMR). After re-making initial settings for the Ethernet module, initialize the transmit/receive descriptors and transmit/receive buffers.
Rev. 6.00 Jun. 12, 2007 Page 312 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
Transmission task
1.
Interrupt handler
Generation of EtherC/E-DMAC interrupt Save EESR and clear the bit by writing 1.
No
Prepare multiple transmit descriptors.
Transmission starts.
2.
After setting the transmit descriptor, set the TR bit in EDTRR to 1.
TC interrupt?
No Yes
Next transmission task generated?
Yes
Make an OS service call to bring the transmission task out of the waiting state. End Interrupt other than TC?
Yes No
Read the TACT bit of the corresponding transmit descriptor.
3.
No TACT = 0? Yes
Interrupt processing for interrupts other than TC
End
4. After setting the corresponding transmit 5. descriptor, set the TR bit in EDTRR to 1.
i = 0;
4.
Call an OS service routine with a timeout function to place the transmission task in a waiting state.
Has the transmission task 1 left the waiting state within * he constant specified time?
Yes
No
5. 6.
Timeout
i++;
i < n? * Yes
2
No 8.
Read the TACT bit of the corresponding transmit descriptor.
Issue a software reset to initialize the EtherC and E-DMAC modules. Make initial settings of the EtherC and E-DMAC modules. Initialize the transmit/receive descriptors and transmit/receive buffers.
7.
TACT = 0? Yes
No
Notes: 1. The specified constant period is the timeout period mentioned in section 12.4.1, Usage Notes on SH-Ether EtherC/E-DMAC Status Register (EESR). 2. Set n with reference to the maximum specified time values in table 12.2. : Processing added as the countermeasure for the problem
Figure 12.13 Countermeasure for the Case with TC Interrupt-Driven Software: Addition of Timeout Processing within the Limit Imposed by the Maximum Specified Time
Rev. 6.00 Jun. 12, 2007 Page 313 of 610 REJ09B0131-0600
Section 12 Ethernet Controller Direct Memory Access Controller (E-DMAC)
Table 12.2 Reference Values for Maximum Specified Time
Communication Rate Maximum Full-duplex with no flow specified time control Half-duplex with no flow control With flow control 10 Mbps 1.3 ms or longer 183 ms or longer (max. 366 ms) 336 ms or longer 100 Mbps 130 s or longer 18.3 ms or longer (max. 36.6 ms) 33.6 ms or longer
Note: The maximum specified time refers to the maximum time taken to transmit a single frame or the maximum time for flow control for a single frame.
Rev. 6.00 Jun. 12, 2007 Page 314 of 610 REJ09B0131-0600
Section 13 Compare Match Timer (CMT)
Section 13 Compare Match Timer (CMT)
This LSI has an on-chip compare match timer (CMT) consisting of a 2-channel 16-bit timer. The CMT has a16-bit counter, and can generate interrupts at set intervals.
13.1
Features
CMT has the following features. * Selection of four counter input clocks Any of four internal clocks (P/8, P/32, P/128, and P/512) can be selected independently for each channel. * Interrupt request on compare match * When not in use, CMT can be stopped by halting its clock supply to reduce power consumption. Figure 13.1 shows a block diagram of CMT.
P/32 P/512 P/128 P/32 P/512 P/128
CMI0
P/8
CMI1
P/8
Control circuit
Clock selection
Control circuit
Clock selection
Comparator
Comparator
CMCOR0
CMCOR1
CMCSR0
CMCSR1
CMCNT0
Channel 0 Module bus CMT [Legend] CMSTR: CMCSR: CMCOR: CMCNT: CMI:
CMCNT1
CMSTR0
Channel 1
Bus interface
Internal bus Compare match timer start register Compare match timer control/status register Compare match timer constant register Compare match counter Compare match interrupt
Figure 13.1 Block Diagram of Compare Match Timer
TIMCMT3A_000020030900
Rev. 6.00 Jun. 12, 2007 Page 315 of 610 REJ09B0131-0600
Section 13 Compare Match Timer (CMT)
13.2
Register Descriptions
The CMT has the following registers. * * * * * * * * Compare match timer start register (CMSTR) Compare match timer control/status register_0 (CMCSR_0) Compare match counter_0 (CMCNT_0) Compare match constant register_0 (CMCOR_0) Compare match timer start register_1 (CMSTR_1) Compare match timer control/status register_1 (CMCSR_1) Compare match counter_1 (CMCNT_1) Compare match constant register_1 (CMCOR_1) Compare Match Timer Start Register (CMSTR)
13.2.1
CMSTR is a 16-bit register that selects whether compare match counter (CMCNT) operates or is stopped. CMSTR is initialized to H'0000 by a power-on reset and a transition to standby mode.
Bit 15 to 2 Bit Name Initial value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 1 STR1 0 R/W Count Start 1 Specifies whether compare match counter 1 operates or is stopped. 0: CMCNT_1 count is stopped 1: CMCNT_1 count is started 0 STR0 0 R/W Count Start 0 Specifies whether compare match counter 0 operates or is stopped. 0: CMCNT_0 count is stopped 1: CMCNT_0 count is started
Rev. 6.00 Jun. 12, 2007 Page 316 of 610 REJ09B0131-0600
Section 13 Compare Match Timer (CMT)
13.2.2
Compare Match Timer Control/Status Register (CMCSR)
CMCSR is a 16-bit register that indicates compare match generation, enables interrupts and selects the counter input clock. CMCSR is initialized to H'0000 by a power-on reset and a transition to standby mode.
Bit 15 to 8 Bit Name Initial value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 7 CMF 0 R/(W)* Compare Match Flag Indicates whether or not the values of CMCNT and CMCOR match. 0: CMCNT and CMCOR values do not match [Clearing condition] When 0 is written to this bit 1: CMCNT and CMCOR values match 6 CMIE 0 R/W Compare Match Interrupt Enable Enables or disables compare match interrupt (CMI) generation when CMCNT and CMCOR values match (CMF=1). 0: Compare match interrupt (CMI) disabled 1: Compare match interrupt (CMI) enabled 5 to 2 All 0 R Reserved These bits are always read as 0. The write value should always be 0.
Rev. 6.00 Jun. 12, 2007 Page 317 of 610 REJ09B0131-0600
Section 13 Compare Match Timer (CMT)
Bit 1 0
Bit Name CKS1 CKS0
Initial value 0 0
R/W R/W R/W
Description Clock Select 1 and 0 Select the clock to be input to CMCNT from four internal clocks obtained by dividing the peripheral operating clock (P). When the STR bit in CMSTR is set to 1, CMCNT starts counting on the clock selected with bits CKS1 and CKS0. 00: P/8 01: P/32 10: P/128 11: P/512
Note: *
Only 0 can be written, to clear the flag.
13.2.3
Compare Match Counter (CMCNT)
CMCNT is a 16-bit register used as an up-counter. When the counter input clock is selected with bits CKS1 and CKS0 in CMCSR and the STR bit in CMSTR is set to 1, CMCNT starts counting using the selected clock. When the value in CMCNT and the value in compare match constant register (CMCOR) match, CMCNT is cleared to H'0000 and the CMF flag in CMCSR is set to 1. CMCNT is initialized to H'0000 by a power-on reset and a transition to standby mode. 13.2.4 Compare Match Constant Register (CMCOR)
CMCOR is a 16-bit register that sets the interval up to a compare match with CMCNT. CMCOR is initialized to H'FFFF by a power-on reset and is initialized to H'FFFF in standby mode.
Rev. 6.00 Jun. 12, 2007 Page 318 of 610 REJ09B0131-0600
Section 13 Compare Match Timer (CMT)
13.3
13.3.1
Operation
Interval Count Operation
When an internal clock is selected with bits CKS1 and CKS0 in CMCSR and the STR bit in CMSTR is set to 1, CMCNT starts incrementing using the selected clock. When the values in CMCNT and CMCOR match, CMCNT is cleared to H'0000 and the CMF flag in CMCSR is set to 1. When the CMIE bit in CMCSR is set to 1, a compare match interrupt (CMI) is requested. CMCNT then starts counting up again from H'0000. Figure 13.2 shows the operation of the compare match counter.
CMCNT1 value
Counter cleared by compare match with CMCOR1
CMCOR1
H'0000
Time
Figure 13.2 Counter Operation 13.3.2 CMCNT Count Timing
One of four internal clocks (P/8, P/32, P/128, and P/512) obtained by dividing the P clock can be selected with bits CKS1 and CKS0 in CMCSR. Figure 13.3 shows the timing.
Peripheral operating clock (P) Nth clock N (N + 1)th clock N+1
Count clock
CMCNT1
Figure 13.3 Count Timing
Rev. 6.00 Jun. 12, 2007 Page 319 of 610 REJ09B0131-0600
Section 13 Compare Match Timer (CMT)
13.4
13.4.1
Interrupts
Interrupt Sources
The CMT has channels and each of them to which a different vector address is allocated has compare match interrupt. When both the interrupt request flag (CMF) and interrupt enable bit (CMIE) are set to 1, the corresponding interrupt request is output. When the interrupt is used to activate a CPU interrupt, the priority of channels can be changed by the interrupt controller settings. For details, see section 6, Interrupt Controller (INTC). 13.4.2 Timing of Setting Compare Match Flag
When CMCOR and CMCNT match, a compare match signal is generated and the CMF bit in CMCSR is set to 1. The compare match signal is generated in the last cycle in which the values match (when the CMCNT value is updated to H'0000). That is, after a match between CMCOR and CMCNT, the compare match signal is not generated until the next CMCNT counter clock input. Figure 13.4 shows the timing of CMF bit setting.
Peripheral operating clock (P)
Counter clock
(N + 1)th clock
CMCNT1
N
0
CMCOR1
N
Compare match signal
Figure 13.4 Timing of CMF Setting 13.4.3 Timing of Clearing Compare Match Flag
The CMF bit in CMCSR is cleared by reading 1 from this bit, then writing 0.
Rev. 6.00 Jun. 12, 2007 Page 320 of 610 REJ09B0131-0600
Section 13 Compare Match Timer (CMT)
13.5
13.5.1
Usage Notes
Conflict between Write and Compare-Match Processes of CMCNT
When the compare match signal is generated in the T2 cycle while writing to CMCNT, clearing CMCNT has priority over writing to it. In this case, CMCNT is not written to. Figure 13.5 shows the timing to clear the CMCNT counter.
CMCSR write cycle T1 Peripheral operating clock (P)
T2
Address
CMCNT
Internal write
Counter clear
CMCNT
N
H'0000
Figure 13.5 Conflict between Write and Compare-Match Processes of CMCNT
Rev. 6.00 Jun. 12, 2007 Page 321 of 610 REJ09B0131-0600
Section 13 Compare Match Timer (CMT)
13.5.2
Conflict between Word-Write and Count-Up Processes of CMCNT
Even when the count-up occurs in the T2 cycle while writing to CMCNT in words, the writing has priority over the count-up. In this case, the count-up is not performed. Figure 13.6 shows the timing to write to CMCNT in words.
CMCSR write cycle T1 Peripheral operating clock (P)
T2
Address
CMCNT
Internal write
CMCNT count-up enable
CMCNT
N
M
Figure 13.6 Conflict between Word-Write and Count-Up Processes of CMCNT
Rev. 6.00 Jun. 12, 2007 Page 322 of 610 REJ09B0131-0600
Section 13 Compare Match Timer (CMT)
13.5.3
Conflict between Byte-Write and Count-Up Processes of CMCNT
Even when the count-up occurs in the T2 cycle while writing to CMCNT in bytes, the byte-writing has priority over the count-up. In this case, the count-up is not performed. The byte data on another side, which is not written to, is also not counted and the previous contents remain. Figure 13.7 shows the timing when the count-up occurs in the T2 cycle while writing to CMCNT in bytes.
CMCSR write cycle T1 Peripheral operating clock (P)
T2
Address
CMCNTH
Internal write
CMCNT count-up enable
CMCNTH
N
M
CMCNTL
X
X
Figure 13.7 Conflict between Byte-Write and Count-Up Processes of CMCNT 13.5.4 Conflict between Write Processes to CMCNT with the Counting Stopped and CMCOR
Writing the same value to CMCNT with the counting stopped and CMCOR is prohibited. If written, the CMF flag in CMCSR is set to 1 and CMCNT is cleared to H'0000.
Rev. 6.00 Jun. 12, 2007 Page 323 of 610 REJ09B0131-0600
Section 13 Compare Match Timer (CMT)
Rev. 6.00 Jun. 12, 2007 Page 324 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Section 14 Serial Communication Interface with FIFO (SCIF)
14.1 Overview
This LSI has a three-channel serial communication interface with FIFO (SCIF) that supports both asynchronous and clock synchronous serial communication. It also has 16-stage FIFO registers for both transmission and reception independently for each channel that enable this LSI to perform efficient high-speed continuous communication. 14.1.1 Features
* Asynchronous serial communication: Serial data communication is performed by start-stop in character units. The SCIF can communicate with a universal asynchronous receiver/transmitter (UART), an asynchronous communication interface adapter (ACIA), or any other communications chip that employs a standard asynchronous serial system. There are eight selectable serial data communication formats. Data length: 7 or 8 bits Stop bit length: 1 or 2 bits Parity: Even, odd, or none Receive error detection: Parity, framing, and overrun errors Break detection: Break is detected when a framing error is followed by at least one frame at the space 0 level (low level). It is also detected by reading the RxD level directly from the port data register when a framing error occurs. * Synchronous mode: Serial data communication is synchronized with a clock signal. The SCIF can communicate with other chips having a synchronous communication function. There is one serial data communication format. Data length: 8 bits Receive error detection: Overrun errors * Full duplex communication: The transmitting and receiving sections are independent, so the SCIF can transmit and receive simultaneously. Both sections use 16-stage FIFO buffering, so high-speed continuous data transfer is possible in both the transmit and receive directions.
SCIS3C4A_000020030900
Rev. 6.00 Jun. 12, 2007 Page 325 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
* On-chip baud rate generator with selectable bit rates * Internal or external transmit/receive clock source: From either baud rate generator (internal) or SCK pin (external) * Four types of interrupts: Transmit-FIFO-data-empty, break, receive-FIFO-data-full, and receive-error interrupts are requested independently. * When the SCIF is not in use, it can be stopped by halting the clock supplied to it, saving power. * In asynchronous, on-chip modem control functions (RTS and CTS) (only for channel 1 and channel 0). * The number of data in the transmit and receive FIFO registers and the number of receive errors of the receive data in the receive FIFO register can be ascertained. * A time-out error (DR) can be detected when receiving in asynchronous mode.
Rev. 6.00 Jun. 12, 2007 Page 326 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Figure 14.1 shows a block diagram of the SCIF for each channel.
Bus interface
Module data bus
Internal data bus
SCSMR SCFRDR (16 stage) SCFTDR (16 stage) SCLSR SCFDR SCFCR RxD SCRSR SCTSR SCFSR SCSCR SCSPTR Transmission/ reception control TxD Parity generation Parity check SCK CTS RTS Clock
SCBRRn
P Baud rate generator P/4 P/16 P/64
External clock TXI RXI ERI BRI SCIF
[Legend] SCRSR: Receive shift register SCFRDR: Receive FIFO data register SCTSR: Transmit shift register SCFTDR: Transmit FIFO data register SCSMR: Serial mode register SCSCR: Serial control register
SCFSR: Serial status register SCBRR: Bit rate register SCSPTR: Serial port register SCFCR: FIFO control register SCFDR: FIFO data count register SCLSR: Line status register
Figure 14.1 Block Diagram of SCIF
Rev. 6.00 Jun. 12, 2007 Page 327 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
14.2
Pin Configuration
The SCIF has the serial pins summarized in table 14.1. Table 14.1 SCIF Pins
Channel 0 Pin Name Serial clock pin Receive data pin Transmit data pin Request to send pin Clear to send pin 1 Serial clock pin Receive data pin Transmit data pin Request to send Clear to send pin 2 Serial clock pin Receive data pin Transmit data pin Abbreviation SCK0 RxD0 TxD0 RTS0 CTS0 SCK1 RxD1 TxD1 RTS1 CTS1 SCK2 RxD2 TxD2 I/O I/O Input Output I/O I/O I/O Input Output Output Input I/O Input Output Function Clock I/O Receive data input Transmit data output Request to send Clear to send Clock I/O Receive data input Transmit data output Request to send Clear to send Clock I/O Receive data input Transmit data output
Rev. 6.00 Jun. 12, 2007 Page 328 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
14.3
Register Description
The SCIF has the following registers. These registers specify the data format and bit rate, and control the transmitter and receiver sections. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Receive FIFO data register_0 (SCFRDR_0) Transmit FIFO data register_0 (SCFTDR_0) Serial mode register_0 (SCSMR_0) Serial control register_0 (SCSCR_0) Serial status register_0 (SCFSR_0) Bit rate register_0 (SCBRR_0) FIFO control register_0 (SCFCR_0) FIFO data count register_0 (SCFDR_0) Serial port register_0 (SCSPTR_0) Line status register_0 (SCLSR_0) Receive FIFO data register_1 (SCFRDR_1) Transmit FIFO data register_1 (SCFTDR_1) Serial mode register_1 (SCSMR_1) Serial control register_1 (SCSCR_1) Serial status register_1 (SCFSR_1) Bit rate register_1 (SCBRR_1) FIFO control register_1 (SCFCR_1) FIFO data count register_1 (SCFDR_1) Serial port register_1 (SCSPTR_1) Line status register_1 (SCLSR_1) Receive FIFO data register_2 (SCFRDR_2) Transmit FIFO data register_2 (SCFTDR_2) Serial mode register_2 (SCSMR_2) Serial control register_2 (SCSCR_2) Serial status register_2 (SCFSR_2) Bit rate register_2 (SCBRR_2) FIFO control register_2 (SCFCR_2) FIFO data count register_2 (SCFDR_2) Serial port register_2 (SCSPTR_2) Line status register_2 (SCLSR_2)
Rev. 6.00 Jun. 12, 2007 Page 329 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
14.3.1
Receive Shift Register (SCRSR)
SCRSR receives serial data. Data input at the RxD pin is loaded into SCRSR in the order received, LSB (bit 0) first, converting the data to parallel form. When one byte has been received, it is automatically transferred to SCFRDR, the receive FIFO data register. The CPU cannot read or write to SCRSR directly. 14.3.2 Receive FIFO Data Register (SCFRDR)
SCFRDR is a 16-stage 8-bit FIFO register that stores serial receive data. The SCIF completes the reception of one byte of serial data by moving the received data from the receive shift register (SCRSR) into SCFRDR for storage. Continuous reception is possible until 16 bytes are stored. The CPU can read but not write to SCFRDR. If data is read when there is no receive data in the SCFRDR, the value is undefined. When this register is full of receive data, subsequent serial data is lost. SCFRDR is initialized to undefined value by a power-on reset.
Bit 7 to 0 Bit Name Initial value R/W Description FIFO for transmits serial data
Undefined R
14.3.3
Transmit Shift Register (SCTSR)
SCTSR transmits serial data. The SCIF loads transmit data from the transmit FIFO data register (SCFTDR) into SCTSR, then transmits the data serially from the TxD pin, LSB (bit 0) first. After transmitting one data byte, the SCIF automatically loads the next transmit data from SCFTDR into SCTSR and starts transmitting again. The CPU cannot read or write to SCTSR directly.
Rev. 6.00 Jun. 12, 2007 Page 330 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
14.3.4
Transmit FIFO Data Register (SCFTDR)
SCFTDR is a 16-stage 8-bit FIFO register that stores data for serial transmission. When the SCIF detects that the transmit shift register (SCTSR) is empty, it moves transmit data written in the SCFTDR into SCTSR and starts serial transmission. Continuous serial transmission is performed until there is no transmit data left in SCFTDR. SCFTDR can always be written to by the CPU. When SCFTDR is full of transmit data (16 bytes), no more data can be written. If writing of new data is attempted, the data is ignored. SCFTDR is initialized to undefined value by a power-on reset.
Bit 7 to 0 Bit Name Initial value R/W Description FIFO for transmits serial data
Undefined W
14.3.5
Serial Mode Register (SCSMR)
SCSMR is a 16-bit register that specifies the SCIF serial communication format and selects the clock source for the baud rate generator. The CPU can always read and write to SCSMR. SCSMR is initialized to H'0000 by a power-on reset.
Bit 15 to 8 Bit Name Initial value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 7 C/A 0 R/W Communication Mode Selects whether the SCIF operates in asynchronous or synchronous mode. 0: Asynchronous mode 1: Synchronous mode
Rev. 6.00 Jun. 12, 2007 Page 331 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Bit 6
Initial Bit Name value CHR 0
R/W R/W
Description Character Length Selects 7-bit or 8-bit data in asynchronous mode. In the synchronous mode, the data length is always eight bits, regardless of the CHR setting. 0: 8-bit data 1: 7-bit data* Note: * When 7-bit data is selected, the MSB (bit 7) of the transmit FIFO data register is not transmitted.
5
PE
0
R/W
Parity Enable Selects whether to add a parity bit to transmit data and to check the parity of receive data, in asynchronous mode. In synchronous mode, a parity bit is neither added nor checked, regardless of the PE setting. 0: Parity bit not added or checked 1: Parity bit added and checked* Note: * When PE is set to 1, an even or odd parity bit is added to transmit data, depending on the parity mode (O/E) setting. Receive data parity is checked according to the even/odd (O/E) mode setting.
4
O/E
0
R/W
Parity mode Selects even or odd parity when parity bits are added and checked. The O/E setting is used only in asynchronous mode and only when the parity enable bit (PE) is set to 1 to enable parity addition and checking. The O/E setting is ignored in synchronous mode, or in asynchronous mode when parity addition and checking is disabled. 0: Even parity*1 1: Odd parity*
2
Notes: 1. If even parity is selected, the parity bit is added to transmit data to make an even number of 1s in the transmitted character and parity bit combined. Receive data is checked to see if it has an even number of 1s in the received character and parity bit combined. 2. If odd parity is selected, the parity bit is added to transmit data to make an odd number of 1s in the transmitted character and parity bit combined. Receive data is checked to see if it has an odd number of 1s in the received character and parity bit combined.
Rev. 6.00 Jun. 12, 2007 Page 332 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Bit 3
Bit Name STOP
Initial value 0
R/W R/W
Description Stop Bit Length Selects one or two bits as the stop bit length in asynchronous mode. This setting is used only in asynchronous mode. It is ignored in synchronous mode because no stop bits are added. When receiving, only the first stop bit is checked, regardless of the STOP bit setting. If the second stop bit is 1, it is treated as a stop bit, but if the second stop bit is 0, it is treated as the start bit of the next incoming character. 0: One stop bit When transmitting, a single 1-bit is added at the end of each transmitted character. 1: Two stop bits When transmitting, two 1 bits are added at the end of each transmitted character.
2
0
R
Reserved This bit is always read as 0. The write value should always be 0.
1 0
CKS1 CKS0
0 0
R/W R/W
Clock Select 1 and 0 Select the internal clock source of the on-chip baud rate generator. Four clock sources are available. P, P/4, P/16 and P/64. For further information on the clock source, bit rate register settings, and baud rate, see section 14.3.8, Bit Rate Register (SCBRR). 00: P 01: P/4 10: P/16 11: P/64 Note: P: Peripheral clock
Rev. 6.00 Jun. 12, 2007 Page 333 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
14.3.6
Serial Control Register (SCSCR)
SCSCR is a 16-bit register that operates the SCIF transmitter/receiver, enables/disables interrupt requests, and selects the transmit/receive clock source. The CPU can always read and write to SCSCR. SCSCR is initialized to H'0000 by a power-on reset.
Bit 15 to 8 Bit Name Initial value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 7 TIE 0 R/W Transmit Interrupt Enable Enables or disables the transmit-FIFO-data-empty interrupt (TXI). Serial transmit data in the transmit FIFO data register (SCFTDR) is send to the transmit shift register (SCTSR). Then, the TDFE flag in the serial status register (SCFSR) is set to1 when the number of data in SCFTDR becomes less than the number of transmission triggers. At this time, a TXI is requested. 0: Transmit-FIFO-data-empty interrupt request (TXI) is disabled* 1: Transmit-FIFO-data-empty interrupt request (TXI) is enabled Note: * The TXI interrupt request can be cleared by writing a greater number of transmit data than the specified transmission trigger number to SCFTDR and by clearing the TDFE bit to 0 after reading 1 from the TDFE bit, or can be cleared by clearing this bit to 0.
Rev. 6.00 Jun. 12, 2007 Page 334 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Bit 6
Bit Name RIE
Initial value 0
R/W R/W
Description Receive Interrupt Enable Enables or disables the receive-data-full (RXI) interrupts requested when the RDF flag or DR flag in serial status register (SCFSR) is set to1, receive-error (ERI) interrupts requested when the ER flag in SCFSR is set to1, and break (BRI) interrupts requested when the BRK flag in SCFSR or the ORER flag in line status register (SCLSR) is set to1. 0: Receive-data-full interrupt (RXI), receive-error interrupt (ERI), and break interrupt (BRI) requests are disabled* 1: Receive-data-full interrupt (RXI), receive-error interrupt (ERI), and break interrupt (BRI) requests are enabled Note: * RXI interrupt requests can be cleared by reading the DR or RDF flag after it has been set to 1, then clearing the flag to 0, or by clearing RIE to 0. ERI or BRI interrupt requests can be cleared by reading the ER, BR or ORER flag after it has been set to 1, then clearing the flag to 0, or by clearing RIE and REIE to 0.
5
TE
0
R/W
Transmit Enable Enables or disables the SCIF serial transmitter. 0: Transmitter disabled 1: Transmitter enabled* Note: * Serial transmission starts after writing of transmit data into SCFTDR. Select the transmit format in SCSMR and SCFCR and reset the transmit FIFO before setting TE to 1.
Rev. 6.00 Jun. 12, 2007 Page 335 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Bit 4
Bit Name RE
Initial value 0
R/W R/W
Description Receive Enable Enables or disables the SCIF serial receiver. 0:Receiver disabled*
1 2
1: Receiver enabled*
Notes: 1. Clearing RE to 0 does not affect the receive flags (DR, ER, BRK, RDF, FER, PER, and ORER). These flags retain their previous values. 2. Serial reception starts when a start bit is detected in asynchronous mode, or synchronous clock input is detected in synchronous mode. Select the receive format in SCSMR and SCFCR and reset the receive FIFO before setting RE to 1. 3 REIE 0 R Receive Error Interrupt Enable Enables or disables the receive-error (ERI) interrupts and break (BRI) interrupts. The setting of REIE bit is valid only when RIE bit is set to 0. 0: Receive-error interrupt (ERI) and break interrupt (BRI) requests are disabled* 1: Receive-error interrupt (ERI) and break interrupt (BRI) requests are enabled Note: * ERI or BRI interrupt requests can be cleared by reading the ER, BR or ORER flag after it has been set to 1, then clearing the flag to 0, or by clearing RIE and REIE to 0. Even if RIE is set to 0, when REIE is set to 1, ERI or BRI interrupt requests are enabled. 2 0 R Reserved This bit is always read as 0. The write value should always be 0.
Rev. 6.00 Jun. 12, 2007 Page 336 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Bit 1 0
Bit Name CKE1 CKE0
Initial value 0 0
R/W R/W R/W
Description Clock Enable 1 and 0 Select the SCIF clock source and enable or disable clock output from the SCK pin. Depending on the combination of CKE1 and CKE0, the SCK pin can be used for serial clock output or serial clock input. The CKE0 setting is valid only when the SCIF is operating on the internal clock (CKE1 = 0). The CKE0 setting is ignored when an external clock source is selected (CKE1 = 1). In synchronous mode, select the SCIF operating mode in the serial mode register (SCSMR), then set CKE1 and CKE0. * Asynchronous mode 00: Internal clock, SCK pin used for input pin (The input signal is ignored. The state of the SCK pin depends on both the SCKIO and SCKDT bits.) 01: Internal clock, SCK pin used for clock output (The output clock frequency is 16 times the bit rate.) 10: External clock, SCK pin used for clock input (The input clock frequency is 16 times the bit rate.) 11: Setting prohibited * Synchronous mode 00: Internal clock, SCK pin used for serial clock output 01: Internal clock, SCK pin used for serial clock output 10: External clock, SCK pin used for serial clock input 11: Setting prohibited
Rev. 6.00 Jun. 12, 2007 Page 337 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
14.3.7
Serial Status Register (SCFSR)
SCFSR is a 16-bit register. The upper 8 bits indicate the number of receives errors in the SCFRDR data, and the lower 8 bits indicate the status flag indicating SCIF operating state. The CPU can always read and write to SCFSR, but cannot write 1 to the status flags (ER, TEND, TDFE, BRK, RDF, and DR). These flags can be cleared to 0 only if they have first been read (after being set to 1). Bits 3 (FER) and 2 (PER) are read-only bits that cannot be written. SCFSR is initialized to H'0060 by a power-on reset.
Bit 15 14 13 12 Bit Name PER3 PER2 PER1 PER0 Initial value 0 0 0 0 R/W R R R R Description Number of Parity Errors Indicate the number of data including a parity error in the receive data stored in the receive FIFO data register (SCFRDR). The value indicated by bits 15 to 12 represents the number of parity errors in SCFRDR. When parity errors have occurred in all 16byte receive data in SCFRDR, PER3 to PER0 show 0. Number of Framing Errors Indicate the number of data including a framing error in the receive data stored in SCFRDR. The value indicated by bits 11 to 8 represents the number of framing errors in SCFRDR. When framing errors have occurred in all 16-byte receive data in SCFRDR, FER3 to FER0 show 0.
11 10 9 8
FER3 FER2 FER1 FER0
0 0 0 0
R R R R
Rev. 6.00 Jun. 12, 2007 Page 338 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Bit 7
Bit Name ER
Initial value 0
R/W
Description
R/(W)* Receive Error Indicates the occurrence of a framing error, or of a 1 parity error when receiving data that includes parity.* 0: Receiving is in progress or has ended normally [Clearing conditions] * * ER is cleared to 0 a power-on reset ER is cleared to 0 when the chip is when 0 is written after 1 is read from ER
1: A framing error or parity error has occurred. [Setting conditions] * ER is set to 1 when the stop bit is 0 after checking whether or not the last stop bit of the received data is 1 at the end of one data receive 2 operation*
*
ER is set to 1 when the total number of 1s in the receive data plus parity bit does not match the even/odd parity specified by the O/E bit in SCSMR Notes: 1. Clearing the RE bit to 0 in SCSCR does not affect the ER bit, which retains its previous value. Even if a receive error occurs, the receive data is transferred to SCFRDR and the receive operation is continued. Whether or not the data read from SCRDR includes a receive error can be detected by the FER and PER bits in SCFSR. 2. In two stop bits mode, only the first stop bit is checked; the second stop bit is not checked.
Rev. 6.00 Jun. 12, 2007 Page 339 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Bit 6
Bit Name TEND
Initial value 0
R/W
Description
R/(W)* Transmit End Indicates that when the last bit of a serial character was transmitted, SCFTDR did not contain valid data, so transmission has ended. 0: Transmission is in progress [Clearing conditions] * TEND is cleared to 0 when 0 is written after 1 is read from TEND after transmit data is written in SCFTDR
1: End of transmission [Setting conditions] * * * TEND is set to 1 when the chip is a power-on reset TEND is set to 1 when TE is cleared to 0 in the serial control register (SCSCR) TEND is set to 1 when SCFTDR does not contain receive data when the last bit of a one-byte serial character is transmitted
Rev. 6.00 Jun. 12, 2007 Page 340 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Bit 5
Bit Name TDFE
Initial value 0
R/W
Description
R/(W)* Transmit FIFO Data Empty Indicates that data has been transferred from the transmit FIFO data register (SCFTDR) to the transmit shift register (SCTSR), the number of data in SCFTDR has become less than the transmission trigger number specified by the TTRG1 and TTRG0 bits in the FIFO control register (SCFCR), and writing of transmit data to SCFTDR is enabled. 0: The number of transmit data written to SCFTDR is greater than the specified transmission trigger number [Clearing conditions] * TDFE is cleared to 0 when data exceeding the specified transmission trigger number is written to SCFTDR after 1 is read from the TDFE bit and then 0 is written TDFE is cleared to 0 when DMAC write data exceeding the specified transmission trigger number to SCFTDR
*
1: The number of transmit data in SCFTDR is equal to or less than the specified transmission trigger number* [Setting conditions] * * TDFE is set to 1 by a power-on reset TDFE is set to 1 when the number of transmit data in SCFTDR has become equal to or less than the specified transmission trigger number as a result of transmission
Note: * Since SCFTDR is a 16-byte FIFO register, the maximum number of data that can be written when TDFE is 1 is "16 minus the specified transmission trigger number". If an attempt is made to write additional data, the data is ignored. The number of data in SCFTDR is indicated by the upper 8 bits of SCFDR.
Rev. 6.00 Jun. 12, 2007 Page 341 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Bit 4
Bit Name BRK
Initial value 0
R/W
Description
R/(W)* Break Detection Indicates that a break signal has been detected in receive data. 0: No break signal received [Clearing conditions] * * BRK is cleared to 0 when the chip is a power-on reset BRK is cleared to 0 when software reads BRK after it has been set to 1, then writes 0 to BRK
1: Break signal received* [Setting condition] * BRK is set to 1 when data including a framing error is received, and a framing error occurs with space 0 in the subsequent receive data
Note: * When a break is detected, transfer of the receive data (H'00) to SCFRDR stops after detection. When the break ends and the receive signal becomes mark 1, the transfer of receive data resumes. 3 FER 0 R Framing Error Indicates a framing error in the data read from the next receive FIFO data register (SCFRDR) in asynchronous mode. 0: No receive framing error occurred in the next data read from SCFRDR [Clearing conditions] * * FER is cleared to 0 when the chip undergoes a power-on reset FER is cleared to 0 when no framing error is present in the next data read from SCFRDR
1: A receive framing error occurred in the next data read from SCFRDR. [Setting condition] * FER is set to 1 when a framing error is present in the next data read from SCFRDR
Rev. 6.00 Jun. 12, 2007 Page 342 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Bit 2
Bit Name PER
Initial value 0
R/W R
Description Parity Error Indicates a parity error in the data read from the next receive FIFO data register (SCFRDR) in asynchronous mode. 0: No receive parity error occurred in the next data read from SCFRDR [Clearing conditions] * * PER is cleared to 0 when the chip undergoes a power-on reset PER is cleared to 0 when no parity error is present in the next data read from SCFRDR
1: A receive parity error occurred in the data read from SCFRDR [Setting condition] * PER is set to 1 when a parity error is present in the next data read from SCFRDR
Rev. 6.00 Jun. 12, 2007 Page 343 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Bit 1
Bit Name RDF
Initial value 0
R/W
Description
R/(W)* Receive FIFO Data Full Indicates that receive data has been transferred to the receive FIFO data register (SCFRDR), and the number of data in SCFRDR has become more than the receive trigger number specified by the RTRG1 and RTRG0 bits in the FIFO control register (SCFCR). 0: The number of transmit data written to SCFRDR is less than the specified receive trigger number [Clearing conditions] * * RDF is cleared to 0 by a power-on reset RDF is cleared to 0 when the SCFRDR is read until the number of receive data in SCFRDR becomes less than the specified receive trigger number after 1 is read from RDF and then 0 is written
1: The number of receive data in SCFRDR is more than the specified receive trigger number [Setting condition] * RDF is set to 1 when a number of receive data more than the specified receive trigger number is stored in SCFRDR*
Note: * SCFTDR is a 16-byte FIFO register. When RDF is 1, the specified receive trigger number of data can be read at the maximum. If an attempt is made to read after all the data in SCFRDR has been read, the data is undefined. The number of receive data in SCFRDR is indicated by the lower 8 bits of SCFDR.
Rev. 6.00 Jun. 12, 2007 Page 344 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Bit 0
Bit Name DR
Initial value 0
R/W
Description
R/(W)* Receive Data Ready Indicates that the number of data in the receive FIFO data register (SCFRDR) is less than the specified receive trigger number, and that the next data has not yet been received after the elapse of 15 ETU from the last stop bit in asynchronous mode. In clock synchronous mode, this bit is not set to 1. 0: Receiving is in progress, or no receive data remains in SCFRDR after receiving ended normally [Clearing conditions] * * DR is cleared to 0 when the chip undergoes a power-on reset DR is cleared to 0 when all receive data are read after 1 is read from DR and then 0 is written
1: Next receive data has not been received [Setting conditions] * DR is set to 1 when SCFRDR contains less data than the specified receive trigger number, and the next data has not yet been received after the elapse of 15 ETU from the last stop bit.*
Note: * This is equivalent to 1.5 frames with the 8-bit, 1-stop-bit format. (ETU: elementary time unit) Note: * The only value that can be written is 0 to clear the flag.
Rev. 6.00 Jun. 12, 2007 Page 345 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
14.3.8
Bit Rate Register (SCBRR)
SCBRR is an 8-bit register that, together with the baud rate generator clock source selected by the CKS1 and CKS0 bits in the serial mode register (SCSMR), determines the serial transmit/receive bit rate. The CPU can always read and write to SCBRR. SCBRR is initialized to H'FF by a power-on reset. Each channel has independent baud rate generator control, so different values can be set in three channels. The SCBRR setting is calculated as follows: * Asynchronous mode:
N= P x 106 - 1 64 x 22n-1 x B
* Synchronous mode:
N= P x 106 - 1 8 x 22n-1 x B
B: N:
Bit rate (bits/s) SCBRR setting for baud rate generator (0 N 255) (The setting value should satisfy the electrical characteristics.) P: Operating frequency for peripheral modules (MHz) n: Baud rate generator clock source (n = 0, 1, 2, 3) (for the clock sources and values of n, see table 14.2.)
Rev. 6.00 Jun. 12, 2007 Page 346 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Table 14.2 SCSMR Settings
SCSMR Settings n 0 1 2 3 Clock Source P P/4 P/16 P/64
P x 106 -1 (N + 1) x B x 642n-1 x 2
CKS1 0 0 1 1
CKS0 0 1 0 1
Note: The bit rate error in asynchronous is given by the following formula:
Error (%) = x 100
Table 14.3 lists examples of SCBRR settings in asynchronous mode, and table 14.4 lists examples of SCBRR settings in synchronous mode. Table 14.3 Bit Rates and SCBRR Settings in Asynchronous Mode
P (MHz) 5 Bit Rate (bits/s) n 110 150 300 600 1200 2400 4800 9600 19200 31250 38400 2 2 1 1 0 0 0 0 0 0 0 N 88 64 129 64 129 64 32 15 7 4 3 Error (%) n -0.25 0.16 0.16 0.16 0.16 0.16 -1.36 1.73 1.73 0.00 1.73 2 2 1 1 0 0 0 0 0 0 0 N 106 77 155 77 155 77 38 19 9 5 4 6 Error (%) n -0.44 0.16 0.16 0.16 0.16 0.16 0.16 -2.34 -2.34 0.00 -2.34 2 2 1 1 0 0 0 0 0 0 0 N 108 79 159 79 159 79 39 19 9 5 4 6.144 Error (%) 0.08 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 2.40 0.00
Rev. 6.00 Jun. 12, 2007 Page 347 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
P (MHz) 7.3728 Bit Rate (bits/s) n 110 150 300 600 1200 2400 4800 9600 19200 31250 38400 2 2 1 1 0 0 0 0 0 0 0 N 130 95 191 95 191 95 47 23 11 6 5 Error (%) n -0.07 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 5.33 0.00 2 2 1 1 0 0 0 0 0 0 0 N 141 103 207 103 207 103 51 25 12 7 6 8 Error (%) n 0.03 0.16 0.16 0.16 0.16 0.16 0.16 0.16 0.16 0.00 -6.99 2 2 1 1 0 0 0 0 0 0 0 N 174 127 255 127 255 127 63 31 15 9 7 9.8304 Error (%) -0.26 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 -1.70 0.00
P (MHz) 10 Bit Rate (bits/s) 110 150 300 600 1200 2400 4800 9600 19200 31250 38400 n 2 2 2 1 1 0 0 0 0 0 0 N 177 129 64 129 64 129 64 32 15 9 7 Error (%) -0.25 0.16 0.16 0.16 0.16 0.16 0.16 -1.36 1.73 0.00 1.73 n 2 2 2 1 1 0 0 0 0 0 0 N 212 155 77 155 77 155 77 38 19 11 9 12 Error (%) 0.03 0.16 0.16 0.16 0.16 0.16 0.16 0.16 0.16 0.00 -2.34 n 2 2 2 1 1 0 0 0 0 0 0 12.288 N 217 159 79 159 79 159 79 39 19 11 9 Error (%) 0.08 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 2.40 0.00 n 3 2 2 1 1 0 0 0 0 0 0 14.7456 N 64 191 95 191 95 191 95 47 23 14 11 Error (%) 0.70 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 -1.70 0.00
Rev. 6.00 Jun. 12, 2007 Page 348 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
P (MHz) 16 Bit Rate (bits/s) 110 150 300 600 1200 2400 4800 9600 19200 31250 38400 n 3 2 2 1 1 0 0 0 0 0 0 N 70 207 103 207 103 207 103 51 25 15 12 Error (%) 0.03 0.16 0.16 0.16 0.16 0.16 0.16 0.16 0.16 0.00 0.16 n 3 2 2 1 1 0 0 0 0 0 0 19.6608 N 86 255 127 255 127 255 127 63 31 19 15 Error (%) 0.31 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 -1.70 0.00 n 3 3 2 2 1 1 0 0 0 0 0 N 88 64 129 64 129 64 129 64 32 19 15 20 Error (%) -0.25 0.16 0.16 0.16 0.16 0.16 0.16 0.16 -1.36 0.00 1.73 n 3 3 2 2 1 1 0 0 0 0 0 N 106 77 155 77 155 77 155 77 38 23 19 24 Error (%) -0.44 0.16 0.16 0.16 0.16 0.16 0.16 0.16 0.16 0.00 -2.34
P (MHz) 24.576 Bit Rate (bits/s) 110 150 300 600 1200 2400 4800 9600 19200 31250 38400 n 3 3 2 2 1 1 0 0 0 0 0 N 108 79 159 79 159 79 159 79 39 24 19 Error (%) 0.08 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 -1.70 0.00 n 3 3 2 2 1 1 0 0 0 0 0 28.7 N 126 92 186 92 186 92 186 92 46 28 22 Error (%) 0.31 0.46 -0.08 0.46 -0.08 0.46 -0.08 0.46 -0.61 -1.03 1.55 n 3 3 2 2 1 1 0 0 0 0 0 N 132 97 194 97 194 97 194 97 48 29 23 30 Error (%) 0.13 -0.35 0.16 -0.35 0.16 -0.35 -1.36 -0.35 -0.35 0.00 1.73 n 3 3 2 2 1 1 0 0 0 0 0 N 145 106 214 106 214 106 214 106 53 32 26 33 Error (%) 0.33 0.39 -0.07 0.39 -0.07 0.39 -0.07 0.39 -0.54 0.00 -0.54
Note: Settings with an error of 1% or less are recommended.
Rev. 6.00 Jun. 12, 2007 Page 349 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Table 14.4 Bit Rates and SCBRR Settings in Synchronous Mode
P (MHz) Bit Rate (bits/s) 110 250 500 1k 2.5k 5k 10k 25k 50k 100k 250k 500k 1M 2M 5 n 3 3 2 1 0 0 0 0 0 N 77 38 77 124 249 124 49 24 4 n 3 2 2 1 1 0 0 0 0 0 0 0 0 8 N 124 249 124 199 99 199 79 39 19 7 3 1 0* n 3 3 2 2 1 1 0 0 0 0 0 0 0 16 N 249 124 249 99 199 99 159 79 39 15 7 3 1 n 3 3 2 2 1 1 0 0 28.7 N 223 111 178 89 178 71 143 71 n 3 3 2 2 1 1 0 0 0 0 30 N 233 116 187 93 187 74 149 74 29 14 n 3 3 2 2 1 1 0 0 0 0 0 33 N 255 125 200 100 200 80 160 80 31 15 7
[Legend] Blank: No setting possible : Setting possible, but error occurs *: Continuous transmission/reception is disabled. Note: Settings with an error of 1% or less are recommended.
Rev. 6.00 Jun. 12, 2007 Page 350 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Table 14.5 indicates the maximum bit rates in asynchronous mode when the baud rate generator is used. Tables 14.6 and 14.7 list the maximum rates for external clock input. Table 14.5 Maximum Bit Rates for Various Frequencies with Baud Rate Generator (Asynchronous Mode)
Settings P (MHz) 5 4.9152 8 9.8304 12 14.7456 16 19.6608 20 24 24.576 28.7 30 33 Maximum Bit Rate (bits/s) 156250 153600 250000 307200 375000 460800 500000 614400 625000 750000 768000 896875 937500 1031250 n 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Rev. 6.00 Jun. 12, 2007 Page 351 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Table 14.6 Maximum Bit Rates with External Clock Input (Asynchronous Mode)
P (MHz) 5 4.9152 8 9.8304 12 14.7456 16 19.6608 20 24 24.576 28.7 30 33 External Input Clock (MHz) 1.2500 1.2288 2.0000 2.4576 3.0000 3.6864 4.0000 4.9152 5.0000 6.0000 6.1440 7.1750 7.5000 8.25 Maximum Bit Rate (bits/s) 78125 76800 125000 153600 187500 230400 250000 307200 312500 375000 384000 448436 468750 515625
Table 14.7 Maximum Bit Rates with External Clock Input (Synchronous Mode)
P (MHz) 5 8 16 24 28.7 30 33 External Input Clock (MHz) 0.8333 1.3333 2.6667 4.0000 4.7833 5.0000 5.5000 Maximum Bit Rate (bits/s) 833333.3 1333333.3 2666666.7 4000000.0 4783333.3 5000000.0 5500000.0
Rev. 6.00 Jun. 12, 2007 Page 352 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
14.3.9
FIFO Control Register (SCFCR)
SCFCR is a 16-bit register that resets the number of data in the transmit and receive FIFO registers, sets the trigger data number, and contains an enable bit for loop-back testing. SCFCR can always be read and written to by the CPU. It is initialized to H'0000 by a power-on reset.
Bit 15 to 11 Bit Name Initial value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 10 9 8 RSTRG2 RSTRG1 RSTRG0 0 0 0 R/W R/W R/W RTS Output Active Trigger When the number of receive data in the receive FIFO register (SCFRDR) becomes more than the number shown below, the RTS signal is set to high. These bits are available only in SCFCR_0 and SCFCR_1. In SCFCR_2, these bits are reserved. The initial value is 0 and the write value should always be 0. 000: 15 001: 1 010: 4 011: 6 100: 8 101: 10 110: 12 111: 14
Rev. 6.00 Jun. 12, 2007 Page 353 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Bit 7 6
Bit Name RTRG1 RTRG0
Initial value 0 0
R/W R/W R/W
Description Receive FIFO Data Trigger Set the specified receive trigger number. The receive data full (RDF) flag in the serial status register (SCFSR) is set when the number of receive data stored in the receive FIFO register (SCFRDR) exceeds the specified trigger number shown below. * Asynchronous mode 00: 1 01: 4 10: 8 11: 14 * Synchronous mode 00: 1 01: 2 10: 8 11: 14
5 4
TTRG1 TTRG0
0 0
R/W R/W
Transmit FIFO Data Trigger 1 and 0 Set the specified transmit trigger number. The transmit FIFO data register empty (TDFE) flag in the serial status register (SCFSR) is set when the number of transmit data in the transmit FIFO data register (SCFTDR) becomes less than the specified trigger number shown below. 00: 8 (8)* 01: 4 (12)* 10: 2 (14)* 11: 0 (16)* Note: * Values in parentheses mean the number of remaining bytes in SCFTDR when the TDFE flag is set to 1.
Rev. 6.00 Jun. 12, 2007 Page 354 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Bit 3
Bit Name MCE
Initial value 0
R/W R/W
Description Modem Control Enable Enables modem control signals CTS and RTS. In synchronous mode, clear this bit to 0. This bit is available only in SCFCR_0 and SCFCR_1. In SCFCR_2, this bit is reserved. The initial value is 0 and the write value should always be 0. 0: Modem signal disabled* 1: Modem signal enabled Note: * The CTS signal is fixed active 0 regardless of the input value, and the RTS signal is also fixed 0.
2
TFRST
0
R/W
Transmit FIFO Data Register Reset Disables the transmit data in the transmit FIFO data register and resets the data to the empty state. 0: Reset operation disabled* 1: Reset operation enabled Note: * Reset operation is executed by a power-on reset.
1
RFRST
0
R/W
Receive FIFO Data Register Reset Disables the receive data in the receive FIFO data register and resets the data to the empty state. 0: Reset operation disabled* 1: Reset operation enabled Note: * Reset operation is executed by a power-on reset.
0
LOOP
0
R/W
Loop-Back Test Internally connects the transmit output pin (TxD) and receive input pin (RxD) and enables loop-back testing. 0: Loop back test disabled 1: Loop back test enabled
Rev. 6.00 Jun. 12, 2007 Page 355 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
14.3.10 FIFO Data Count Register (SCFDR) SCFDR is a 16-bit register which indicates the number of data stored in the transmit FIFO data register (SCFTDR) and the receive FIFO data register (SCFRDR). It indicates the number of transmit data in SCFTDR with the upper eight bits, and the number of receive data in SCFRDR with the lower eight bits. SCFDR can always be read from by the CPU. SCFDR is initialized to H'0000 by a power on reset.
Bit 15 to 13 Bit Name Initial value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 12 11 10 9 8 7 to 5 T4 T3 T2 T1 T0 0 0 0 0 0 All 0 R R R R R R Reserved These bits are always read as 0. The write value should always be 0. 4 3 2 1 0 R4 R3 R2 R1 R0 0 0 0 0 0 R R R R R Indicate the number of receive data stored in SCFRDR. H'00 means no receive data, and H'10 means that SCFRDR full of receive data. Indicate the number of non-transmitted data stored in SCFTDR. H'00 means no transmit data, and H'10 means that SCFTDR is full of transmit data.
Rev. 6.00 Jun. 12, 2007 Page 356 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
14.3.11 Serial Port Register (SCSPTR) SCSPTR is a 16-bit register that controls input/output and data for the pins multiplexed to the SCIF function. Bits 7 and 6 can control the RTS pin, bits 5 and 4 can control the CTS pin, and bits 3 and 2 can control the SCK pin. Bits 1 and 0 can be used to read the input data from the RxD pin and to output data to the TxD pin, so they control break of serial transfer. In addition to descriptions of individual bits shown below, see section 14.6, Serial Port Register (SCSPTR) and SCIF Pins. SCSPTR can always be read from or written to by the CPU. Bits 7, 5, 3, and 1 in SCSPTR are initialized by a power-on reset.
Bit 15 to 8 Bit Name Initial value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 7 RTSIO 0 R/W RTS Port Input/Output Control Controls the RTS pin in combination with the RTSDT bit in this register and the MCE bit in SCFCR. This bit is reserved in SCPTR_2 of SCIF channel 2 since SCIF channel 2 does not support the flow control. 6 RTSDT * R/W RTS Port Data Controls the RTS pin in combination with the RTSIO bit in this register and the MCE bit in SCFCR. Select the RTS pin function in the PFC (pin function controller) beforehand. MCE RTSIO RTSDT: RTS pin state 0 0 0 1 0 1 1 x x: 0: 1: x: Input (initial state) Low level output High level output Sequence output according to modem control logic
x: Don't care The RTS pin state is read from this bit instead of the set value. This bit is reserved in SCPTR_2 of SCIF channel 2 since SCIF channel 2 does not support the flow control.
Rev. 6.00 Jun. 12, 2007 Page 357 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Bit 5
Bit Name CTSIO
Initial value 0
R/W R/W
Description CTS Port Input/Output Control Controls the CTS pin in combination with the CTSDT bit in this register and the MCE bit in SCFCR. This bit is reserved in SCPTR_2 of SCIF channel 2 since SCIF channel 2 does not support the flow control.
4
CTSDT
*
R/W
CTS Port Data Controls the CTS pin in combination with the CTSIO bit in this register and the MCE bit in SCFCR. Select the CTS pin function in the PFC (pin function controller) beforehand. MCE 0 0 0 1 CTSIO CTSDT: CTS pin state 0 1 1 x x: 0: 1: x: Input (initial state) Low level output High level output Input to modem control logic
x: Don't care The CTS pin state is read from this bit instead of the set value. This bit is reserved in SCPTR_2 of SCIF channel 2 since SCIF channel 2 does not support the flow control. 3 SCKIO 0 R/W SCK Port Input/Output Control Controls the SCK pin in combination with the SCKDT bit in this register, the C/A bit in SCSMR, and bits CKE1 and CKE0 in SCSCR.
Rev. 6.00 Jun. 12, 2007 Page 358 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Bit 2
Bit Name SCKDT
Initial value *
R/W R/W
Description SCK Port Data Controls the SCK pin in combination with the SCKIO bit in this register, the C/A bit in SCSMR, and bits CKE1 and CKE0 in SCSCR. Select the SCK pin function in the PFC (pin function controller) beforehand.
C/A 0 0 0 0 CKE1 0 0 0 0 CKE0 0 0 0 1 SCKIO 0 0 1 x SCKDT: x: 0: 1: x: SCK pin state Input (initial state) Low level output High level output Internal clock output according to serial core logic External clock input to serial core logic Setting prohibited Internal clock output according to serial core logic Internal clock output according to serial core logic External clock input to serial core logic Setting prohibited
0 0 1
1 1 0
0 1 0
x x x
x: x: x:
1
0
1
x
x:
1 1
1 1
0 1
x x
x: x:
x: Don't care The SCK pin state is read from this bit instead of the set value. 1 SPBIO 0 R/W Serial Port Break Input/Output Control Controls the TxD pin in combination with the SPBDT bit in this register and the TE bit in SCSCR.
Rev. 6.00 Jun. 12, 2007 Page 359 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Bit 0
Bit Name SPBDT
Initial value *
R/W R/W
Description Serial Port Break Data Controls the TxD pin in combination with the SPBIO bit in this register and the TE bit in SCSCR. The RxD pin state can also be monitored. Select the TxD or RxD pin function in the PFC (pin function controller) beforehand. TE 0 0 0 0 SPBIO 0 1 1 x SPBDT: x: 0: 1: x: TxD pin state Input (initial state) Low level output High level output Transmit data output according to serial core logic
x: Don't care The RxD pin state is read from this bit instead of the set value. Note: * This bit is read as an undefined value and the setting value is 0.
Rev. 6.00 Jun. 12, 2007 Page 360 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
14.3.12 Line Status Register (SCLSR) SCLSR is a 16-bit readable/writable register which can always be read from and written to by the CPU. However, a 1 cannot be written to the ORER flag. This flag can be cleared to 0 only if it has first been read (after being set to 1). SCLSR is initialized to H'0000 by a power-on reset.
Bit 15 to 1 Bit Name Initial value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 0 ORER 0 R/(W)* Overrun Error Indicates the occurrence of an overrun error. 0: Receiving is in progress or has ended normally * [Clearing conditions] * * ORER is cleared to 0 when the chip is a power-on reset ORER is cleared to 0 when 0 is written after 1 is read from ORER.
2 1
1: An overrun error has occurred * [Setting condition] *
ORER is set to 1 when the next serial receiving is finished while receive FIFO data are full. Notes: 1. Clearing the RE bit to 0 in SCSCR does not affect the ORER bit, which retains its previous value. 2. The receive FIFO data register (SCFRDR) hold the data before an overrun error is occurred, and the next receive data is extinguished. When ORER is set to 1, SCIF can not continue the next serial receiving. Note: * The only value that can be written is 0 to clear the flag.
Rev. 6.00 Jun. 12, 2007 Page 361 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
14.4
14.4.1
Operation
Overview
For serial communication, the SCIF has an asynchronous mode in which characters are synchronized individually, and a synchronous mode in which communication is synchronized with clock pulses. The SCIF has a 16-byte FIFO buffer for both transmit and receive operations, reducing the overhead of the CPU, and enabling continuous high-speed communication. Moreover, it has RTS and CTS signals as modem control signals (for channels 0 and 1). The transmission format is selected in the serial mode register (SCSMR). The SCIF clock source is selected by the combination of the CKE1 and CKE0 bits in the serial control register (SCSCR). Asynchronous Mode: * Data length is selectable: 7 or 8 bits. * Parity bit is selectable. So is the stop bit length (1 or 2 bits). The combination of the preceding selections constitutes the communication format and character length. * In receiving, it is possible to detect framing errors, parity errors, receive FIFO data full, overrun errors, receive data ready, and breaks. * The number of stored data bytes is indicated for both the transmit and receive FIFO registers. * An internal or external clock can be selected as the SCIF clock source. When an internal clock is selected, the SCIF operates using the on-chip baud rate generator. When an external clock is selected, the external clock input must have a frequency 16 times the bit rate. (The on-chip baud rate generator is not used.) Synchronous Mode: * The transmission/reception format has a fixed 8-bit data length. * In receiving, it is possible to detect overrun errors (ORER). * An internal or external clock can be selected as the SCIF clock source. When an internal clock is selected, the SCIF operates using the on-chip baud rate generator, and outputs a serial clock signal to external devices. When an external clock is selected, the SCIF operates on the input serial clock. The onchip baud rate generator is not used.
Rev. 6.00 Jun. 12, 2007 Page 362 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Table 14.8 SCSMR Settings and SCIF Communication Formats
SCSMR Settings Bit 7 Bit 6 Bit 5 Bit 3 C/A CHR PE STOP Mode 0 0 0 0 1 1 0 1 1 0 0 1 1 0 1 1 Note: * * * * Synchronous 8-bit Not set : Don't care Set 7-bit Not set Set Asynchronous SCIF Communication Format Data Length 8-bit Parity Bit Not set Stop Bit Length 1 bit 2 bits 1 bit 2 bits 1 bit 2 bits 1 bit 2 bits None
Table 14.9 SCSMR and SCSCR Settings and SCIF Clock Source Selection
SCSMR SCSCR Settings Bit 7 C/A 0 Bit 1 CKE1 0 Bit 0 CKE0 0 Mode Clock Source SCIF Transmit/Receive Clock SCK Pin Function SCIF does not use the SCK pin. The state of the SCK pin depends on both the SCKIO and SCKDT bits. Clock with a frequency 16 times the bit rate is output. External Input a clock with frequency 16 times the bit rate. Synchronous Internal Setting prohibited. Serial clock is output.
Asynchronous Internal
1 1 0 1 1 0 1 Note: * * 0 1 : Don't care
External Input the serial clock. Setting prohibited.
Rev. 6.00 Jun. 12, 2007 Page 363 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
14.4.2
Operation in Asynchronous Mode
In asynchronous mode, each transmitted or received character begins with a start bit and ends with a stop bit. Serial communication is synchronized one character at a time. The transmitting and receiving sections of the SCIF are independent, so full duplex communication is possible. The transmitter and receiver are 16-byte FIFO buffered, so data can be written and read while transmitting and receiving are in progress, enabling continuous transmitting and receiving. Figure 14.2 shows the general format of asynchronous serial communication. In asynchronous serial communication, the communication line is normally held in the mark (high) state. The SCIF monitors the line and starts serial communication when the line goes to the space (low) state, indicating a start bit. One serial character consists of a start bit (low), data (LSB first), parity bit (high or low), and stop bit (high), in that order. When receiving in asynchronous mode, the SCIF synchronizes at the falling edge of the start bit. The SCIF samples each data bit on the eighth pulse of a clock with a frequency 16 times the bit rate. Receive data is latched at the center of each bit.
Idle state (mark state) 1 0/1 Parity bit 1 bit or none 1 1
1 Serial data 0 Start bit 1 bit
LSB D0 D1 D2 D3 D4 D5 D6
MSB D7
Stop bit
Transmit/receive data 7 or 8 bits One unit of transfer data (character or frame)
1 or 2 bits
Figure 14.2 Example of Data Format in Asynchronous Communication (8-Bit Data with Parity and Two Stop Bits)
Rev. 6.00 Jun. 12, 2007 Page 364 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Transmit/Receive Formats: Table 14.10 lists the eight communication formats that can be selected in asynchronous mode. The format is selected by settings in the serial mode register (SCSMR). Table 14.10 Serial Communication Formats (Asynchronous Mode)
SCSMR Bits CHR 0 0 0 0 1 1 1 1 PE STOP 0 0 1 1 0 0 1 1 0 1 0 1 0 1 0 1 1 START START START START START START START START 2 Serial Transmit/Receive Format and Frame Length 3 4 5 6 7 8 9 10 STOP STOP STOP P P STOP STOP STOP P P STOP STOP STOP STOP STOP STOP 11 12
8-bit data 8-bit data 8-bit data 8-bit data 7-bit data 7-bit data 7-bit data 7-bit data
[Legend] START: Start bit STOP: Stop bit P: Parity bit
Clock: An internal clock generated by the on-chip baud rate generator or an external clock input from the SCK pin can be selected as the SCIF transmit/receive clock. The clock source is selected by the C/A bit in the serial mode register (SCSMR) and bits CKE1 and CKE0 in the serial control register (SCSCR) (table 14.9). When an external clock is input at the SCK pin, it must have a frequency equal to 16 times the desired bit rate. When the SCIF operates on an internal clock, it can output a clock signal at the SCK pin. The frequency of this output clock is equal to 16 times the desired bit rate.
Rev. 6.00 Jun. 12, 2007 Page 365 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Transmitting and Receiving Data: SCIF Initialization (Asynchronous Mode): Before transmitting or receiving, clear the TE and RE bits to 0 in the serial control register (SCSCR), then initialize the SCIF as follows. When changing the operation mode or the communication format, always clear the TE and RE bits to 0 before following the procedure given below. Clearing TE to 0 initializes the transmit shift register (SCTSR). Clearing TE and RE to 0, however, does not initialize the serial status register (SCFSR), transmit FIFO data register (SCFTDR), or receive FIFO data register (SCFRDR), which retain their previous contents. Clear TE to 0 after all transmit data has been transmitted and the TEND flag in the SCFSR is set. The TE bit can be cleared to 0 during transmission, but the transmit data goes to the Mark state after the bit is cleared to 0. Set the TFRST bit in SCFCR to 1 and reset SCFTDR before TE is set again to start transmission. When an external clock is used, the clock should not be stopped during initialization or subsequent operation. SCIF operation becomes unreliable if the clock is stopped.
Rev. 6.00 Jun. 12, 2007 Page 366 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Figure 14.3 shows a sample flowchart for initializing the SCIF.
Start of initialization Clear TE and RE bits in SCSCR to 0 Set TFRST and RFRST bits in SCFCR to 1 After reading BRK, DR, and ER flags in SCFSR, and each flag in SCLSR, write 0 to clear them Set CKE1 and CKE0 bits in SCSCR (leaving TE, RE, TIE, and RIE bits cleared to 0) Set data transfer format in SCSMR Set value in SCBRR Wait 1-bit interval elapsed? Yes Set RTRG1-0 and TTRG1-0 bits in SCFCR, and clear TFRST and RFRST bits to 0 Set TE and RE bits in SCSCR to 1, and set TIE, RIE, and REIE bits End of initialization [4] No [1]
[1] Set the clock selection in SCSCR. Be sure to clear bits TIE, RIE, TE, and RE to 0. [2] Set the data transfer format in SCSMR. [3] Write a value corresponding to the bit rate into SCBRR. (Not necessary if an external clock is used.) [4] Wait at least one bit interval, then set the TE bit or RE bit in SCSCR to 1. Also set the RIE, REIE, and TIE bits. Setting the TE bit enables the TxD and RxD pins to be used. When transmitting, the SCIF will go to the mark state; when receiving, it will go to the idle state, waiting for a start bit.
[2] [3]
Figure 14.3 Sample Flowchart for SCIF Initialization
Rev. 6.00 Jun. 12, 2007 Page 367 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Transmitting Serial Data (Asynchronous Mode) Figure 14.4 shows a sample flowchart for serial transmission. Use the following procedure for serial data transmission after enabling the SCIF for transmission.
Start of transmission [1] SCIF status check and transmit data write: Read SCFSR and check that the TDFE flag is set to 1, then write transmit data to SCFTDR, and read 1 from the TDFE and TEND flags, then clear to 0. [1] The number of transmit data bytes that can be written is 16 - (transmit trigger set number). [2] Serial transmission continuation procedure: No [2] To continue serial transmission, read 1 from the TDFE flag to confirm that writing is possible, then write data to SCFTDR, and then clear the TDFE flag to 0. [3] Break output at the end of serial transmission: No To output a break in serial transmission, clear the SPBDT bit to 0 and set the SPBIO bit to 1 in SCSPTR, then clear the TE bit in SCSCR to 0. In [1] and [2], it is possible to ascertain the number of data bytes that can be written from the number of transmit data bytes in SCFTDR indicated by the upper 8 bits of SCFDR.
Read TDFE flag in SCFSR No
TDFE = 1? Yes Write transmit data in SCFTDR, and read 1 from TDFE flag and TEND flag in SCFSR, then clear to 0
All data transmitted? Yes Read TEND flag in SCFSR
TEND = 1? Yes Break output? Yes Clear SPBDT to 0 and set SPBIO to 1 Clear TE bit in SCSCR to 0
No
[3]
End of transmission
Figure 14.4 Sample Flowchart for Transmitting Serial Data
Rev. 6.00 Jun. 12, 2007 Page 368 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
In serial transmission, the SCIF operates as described below. 1. When data is written into the transmit FIFO data register (SCFTDR), the SCIF transfers the data from SCFTDR to the transmit shift register (SCTSR) and starts transmitting. Confirm that the TDFE flag in the serial status register (SCFSR) is set to 1 before writing transmit data to SCFTDR. The number of data bytes that can be written is (16 - transmit trigger setting). 2. When data is transferred from SCFTDR to SCTSR and transmission is started, consecutive transmit operations are performed until there is no transmit data left in SCFTDR. When the number of transmit data bytes in SCFTDR falls below the transmit trigger number set in the FIFO control register (SCFCR), the TDFE flag is set. If the TIE bit in the serial control register (SCSR) is set to 1 at this time, a transmit-FIFO-data-empty interrupt (TXI) request is generated. The serial transmit data is sent from the TxD pin in the following order. A. Start bit: One-bit 0 is output. B. Transmit data: 8-bit or 7-bit data is output in LSB-first order. C. Parity bit: One parity bit (even or odd parity) is output. (A format in which a parity bit is not output can also be selected.) D. Stop bit(s): One or two 1 bits (stop bits) are output. E. Mark state: 1 is output continuously until the start bit that starts the next transmission is sent. 3. The SCIF checks the SCFTDR transmit data at the timing for sending the stop bit. If data is present, the data is transferred from SCFTDR to SCTSR, the stop bit is sent, and then serial transmission of the next frame is started. If there is no transmit data, the TEND flag in SCFSR is set to 1, the stop bit is sent, and then the line goes to the mark state in which 1 is output continuously.
Rev. 6.00 Jun. 12, 2007 Page 369 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Figure 14.5 shows an example of the operation for transmission.
Start bit 0 D0 D1 Data D7 Parity Stop Start bit bit bit 0/1 1 0 D0 D1 Data D7 Parity Stop bit bit 0/1 1
1 Serial data
1 Idle state (mark state)
TDFE
TEND TXI interrupt request Data written to SCFTDR and TDFE flag read as 1 then cleared to 0 by TXI interrupt handler One frame TXI interrupt request
Figure 14.5 Example of Transmit Operation (8-Bit Data, Parity, One Stop Bit) 4. When modem control is enabled, transmission can be stopped and restarted in accordance with the CTS input value. When CTS is set to 1, if transmission is in progress, the line goes to the mark state after transmission of one frame. When CTS is set to 0, the next transmit data is output starting from the start bit. Figure 14.6 shows an example of the operation when modem control is used (only for channel 0).
Start bit Serial data TxD 0 D0 D1 Parity Stop bit bit D7 0/1 1 Start bit 0 D0 D1 D7 0/1
CTS
Drive high before stop bit
Figure 14.6 Example of Operation Using Modem Control (CTS)
Rev. 6.00 Jun. 12, 2007 Page 370 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Receiving Serial Data (Asynchronous Mode): Figures 14.7 and 14.8 show a sample flowchart for serial reception. Use the following procedure for serial data reception after enabling the SCIF for reception.
[1] Receive error handling and break detection: Read the DR, ER, and BRK flags in SCFSR, and the ORER flag in SCLSR, to identify any error, perform the appropriate error handling, then clear the DR, ER, BRK, and ORER flags to 0. In the case of a framing error, a break can also be detected by reading the value of the RxD pin. [2] SCIF status check and receive data read: Read SCFSR and check that RDF = 1, then read the receive data in SCFRDR, read 1 from the RDF flag, and then clear the RDF flag to 0. The transition of the RDF flag from 0 to 1 can also be identified by an RXI interrupt. [3] Serial reception continuation procedure: All data received? Yes Clear RE bit in SCSCR to 0 End of reception [3] To continue serial reception, read at least the receive trigger set number of receive data bytes from SCFRDR, read 1 from the RDF flag, then clear the RDF flag to 0. The number of receive data bytes in SCFRDR can be ascertained by reading from SCRFDR.
Start of reception
Read ER, DR, BRK flags in SCFSR and ORER flag in SCLSR
[1]
ER, DR, BRK or ORER = 1? No Read RDF flag in SCFSR No
Yes
Error handling [2]
RDF = 1? Yes Read receive data in SCFRDR, and clear RDF flag in SCFSR to 0
No
Figure 14.7 Sample Flowchart for Receiving Serial Data
Rev. 6.00 Jun. 12, 2007 Page 371 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Error handling No ORER = 1? Yes Overrun error handling [1] Whether a framing error or parity error has occurred in the receive data that is to be read from SCFRDR can be ascertained from the FER and PER bits in SCFSR. [2] When a break signal is received, receive data is not transferred to SCFRDR while the BRK flag is set. However, note that the last data in SCFRDR is H'00, and the break data in which a framing error occurred is stored.
No ER = 1? Yes Receive error handling
No BRK = 1? Yes Break handling
No DR = 1? Yes Read receive data in SCFRDR
Clear DR, ER, BRK flags in SCFSR, and ORER flag in SCLSR, to 0
End
Figure 14.8 Sample Flowchart for Receiving Serial Data (cont)
Rev. 6.00 Jun. 12, 2007 Page 372 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
In serial reception, the SCIF operates as described below. 1. The SCIF monitors the transmission line, and if a 0 start bit is detected, performs internal synchronization and starts reception. 2. The received data is stored in SCRSR in LSB-to-MSB order. 3. The parity bit and stop bit are received. After receiving these bits, the SCIF carries out the following checks. A. Stop bit check: The SCIF checks whether the stop bit is 1. If there are two stop bits, only the first is checked. B. The SCIF checks whether receive data can be transferred from the receive shift register (SCRSR) to SCFRDR. C. Overrun check: The SCIF checks that the ORER flag is 0, indicating that the overrun error has not occurred. D. Break check: The SCIF checks that the BRK flag is 0, indicating that the break state is not set. If all the above checks are passed, the receive data is stored in SCFRDR. Note: When a parity error or a framing error occurs, reception is not suspended. 4. If the RIE bit in SCSCR is set to 1 when the RDF or DR flag changes to 1, a receive-FIFOdata-full interrupt (RXI) request is generated. If the RIE bit or the REIE bit in SCSCR is set to 1 when the ER flag changes to 1, a receive-error interrupt (ERI) request is generated. If the RIE bit or the REIE bit in SCSCR is set to 1 when the BRK or ORER flag changes to 1, a break reception interrupt (BRI) request is generated.
Rev. 6.00 Jun. 12, 2007 Page 373 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Figure 14.9 shows an example of the operation for reception.
Start bit 0 D0 D1 Data D7 Parity Stop Start bit bit bit 0/1 1 0 D0 D1 Data D7 Parity Stop bit bit 0/1 0 0/1
1 Serial data
RDF
FER RXI interrupt request One frame
Data read and RDF flag read as 1 then cleared to 0 by RXI interrupt handler
ERI interrupt request generated by receive error
Figure 14.9 Example of SCIF Receive Operation (8-Bit Data, Parity, One Stop Bit) 5. When modem control is enabled, the RTS signal is output depending on the empty status of SCFRDR. When RTS is 0, reception is possible. When RTS is 1, this indicates that the SCFRDR is full and no extra data can be received. (Only for channel 0 and channel 1) Figure 14.10 shows an example of the operation when modem control is used.
Start bit Serial data RxD 0 D0 D1 D2 Parity Stop bit bit D7 0/1 1 Start bit 0 D0 D1 D7 0/1
RTS
Figure 14.10 Example of Operation Using Modem Control (RTS)
Rev. 6.00 Jun. 12, 2007 Page 374 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
14.4.3
Synchronous Mode
In synchronous mode, the SCIF transmits and receives data in synchronization with clock pulses. This mode is suitable for high-speed serial communication. The SCIF transmitter and receiver are independent, so full-duplex communication is possible while sharing the same clock. The transmitter and receiver are also 16-byte FIFO buffered, so continuous transmitting or receiving is possible by reading or writing data while transmitting or receiving is in progress. Figure 14.11 shows the general format in synchronous serial communication.
One unit of transfer data (character or frame) * Synchronization clock LSB Serial data Don't care Note: * High except in continuous transfer Bit 0 Bit 1 Bit 2 Bit 3 Bit 4 Bit 5 Bit 6 MSB Bit 7 Don't care *
Figure 14.11 Data Format in Synchronous Communication In synchronous serial communication, each data bit is output on the communication line from one falling edge of the serial clock to the next. Data is guaranteed valid at the rising edge of the serial clock. In each character, the serial data bits are transmitted in order from the LSB (first) to the MSB (last). After output of the MSB, the communication line remains in the state of the MSB. In synchronous mode, the SCIF transmits data by synchronizing with the falling edge of the serial clock, and receives data by synchronizing with the rising edge of the serial clock.
Rev. 6.00 Jun. 12, 2007 Page 375 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Communication Format: The data length is fixed at eight bits. No parity bit can be added. Clock: An internal clock generated by the on-chip baud rate generator or an external clock input from the SCK pin can be selected as the SCIF transmit/receive clock. When the SCIF operates on an internal clock, it outputs the clock signal at the SCK pin. Eight clock pulses are output per transmitted or received character. When the SCIF is not transmitting or receiving, the clock signal remains in the high state. When only receiving, the clock signal outputs while the RE bit of SCSCR is 1 and the number of data in receive FIFO is less than the receive FIFO data trigger number. In this case, 8 x (16 + 1) = 136 pulses of synchronous clock are output. To perform reception of n characters of data, select an external clock as the clock source. If an internal clock should be used, set RE = 1 and TE = 1 and receive n characters of data simultaneously with the transmission of n characters of dummy data. Transmitting and Receiving Data SCIF Initialization (Synchronous Mode): Before transmitting, receiving, or changing the mode or communication format, the software must clear the TE and RE bits to 0 in the serial control register (SCSCR), then initialize the SCIF. Clearing TE to 0 initializes the transmit shift register (SCTSR). Clearing RE to 0, however, does not initialize the RDF, PER, FER, and ORER flags and receive data register (SCRDR), which retain their previous contents. Figure 14.12 shows a sample flowchart for initializing the SCIF. The procedure for initializing the SCIF is:
Rev. 6.00 Jun. 12, 2007 Page 376 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Start of initialization
Clear TE and RE bits in SCSCR to 0 Set TFRST and RFRST bits in SCFCR to 1 to clear the FIFO buffer After reading BRK, DR, and ER flags in SCFSR and a flag in SCLSR, write 0 to clear them Set CKE1 and CKE0 bits in SCSCR (leaving TE, RE, TIE, and RIE bits cleared to 0) Set data transfer format in SCSMR Set value in SCBRR Wait 1-bit interval elapsed? Yes Set RTRG1-0 and TTRG1-0 bits in SCFCR, and clear TFRST and RFRST bits to 0 Set TE and RE bits in SCSCR to 1, and set TIE, RIE, and REIE bits End of initialization
[1]
[1] Leave the TE and RE bits cleared to 0 until the initialization almost ends. Be sure to clear the TIE, RIE, TE, and RE bits to 0. [2] Set the CKE1 and CKE0 bits. [3] Set the data transfer format in SCSMR. [4] Write a value corresponding to the bit rate into SCBRR. This is not necessary if an external clock is used. Wait at least one bit interval after this write before moving to the next step.
[2]
[3] [4]
No
[5] Set the TE or RE bit in SCSCR to 1. Also set the TEI, RIE, and REIE bits to enable the TxD, RxD, and SCK pins to be used. When transmitting, the TxD pin will go to the mark state. When receiving in clocked synchronous mode with the synchronization clock output (clock master) selected, a clock starts to be output from the SCIF_CLK pin at this point.
[5]
Figure 14.12 Sample Flowchart for SCIF Initialization
Rev. 6.00 Jun. 12, 2007 Page 377 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Transmitting Serial Data (Synchronous Mode): Figure 14.13 shows a sample flowchart for transmitting serial data.
Start of transmission [1] SCIF status check and transmit data write: No Read SCFSR and check that the TDFE flag is set to 1, then write transmit data to SCFTDR. Read the TDFE and TEND flags while they are 1, then clear them to 0. [1]
Read TDFE flag in SCFSR
TDFE = 1? Yes Write transmit data to SCFTDR Read TDFE and TEND flags in SCFSR while they are 1, then clear them to 0
[2] Serial transmission continuation procedeure: To continue serial transmission, read 1 from the TDFE flag to confirm that writing is possible, them write data to SCFTDR, and then clear the TDFE flag to 0.
All data transmitted? [2] Yes Read TEND flag in SCFSR
No
TEND = 1? Yes Clear TE bit in SCSCR to 0
No
End of transmission
Figure 14.13 Sample Flowchart for Transmitting Serial Data
Rev. 6.00 Jun. 12, 2007 Page 378 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
In transmitting serial data, the SCIF operates as follows: 1. When data is written into the transmit FIFO data register (SCFTDR), the SCIF transfers the data from SCFTDR to the transmit shift register (SCTSR) and starts transmitting. Confirm that the TDFE flag in the serial status register (SCFSR) is set to 1 before writing transmit data to SCFTDR. The number of data bytes that can be written is (16 - transmit trigger setting). 2. When data is transferred from SCFTDR to SCTSR and transmission is started, consecutive transmit operations are performed until there is no transmit data left in SCFTDR. When the number of transmit data bytes in SCFTDR falls below the transmit trigger number set in the FIFO control register (SCFCR), the TDFE flag is set. If the TIE bit in the serial control register (SCSR) is set to 1 at this time, a transmit-FIFO-data-empty interrupt (TXI) request is generated. If clock output mode is selected, the SCIF outputs eight synchronous clock pulses. If an external clock source is selected, the SCIF outputs data in synchronization with the input clock. Data is output from the TxD pin in order from the LSB (bit 0) to the MSB (bit 7). 3. The SCIF checks the SCFTDR transmit data at the timing for sending the MSB (bit 7). If data is present, the data is transferred from SCFTDR to SCTSR, the MSB (bit 7) is sent, and then serial transmission of the next frame is started. If there is no transmit data, the TEND flag in SCFSR is set to 1, the MSB (bit 7) is sent, and then the TxD pin holds the states. 4. After the end of serial transmission, the SCK pin is held in the high state. Figure 14.14 shows an example of SCIF transmit operation.
Synchronization clock Serial data LSB Bit 0 Bit 1 MSB Bit 7 Bit 0 Bit 1 Bit 6 Bit 7
TDFE
TEND
TXI interrupt request Data written to SCFTDR TXI and TDFE flag cleared interrupt to 0 by TXI interrupt request handler
One frame
Figure 14.14 Example of SCIF Transmit Operation
Rev. 6.00 Jun. 12, 2007 Page 379 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Receiving Serial Data (Synchronous Mode): Figure 14.15 shows a sample flowchart for receiving serial data. When switching from asynchronous mode to synchronous mode without SCIF initialization, make sure that ORER, PER, and FER are cleared to 0.
[1] Receive error handling: Read the ORER flag in SCLSR to identify any error, perform the appropriate error handling, then clear the ORER flag to 0. Transmission/reception cannot be resumed while the ORER flag is set to 1. Yes [1] No Read RDF flag in SCFSR No Error handling [2] [2] SCIF status check and receive data read: Read SCFSR and check that RDF = 1, then read the receive data in SCFRDR, and clear the RDF flag to 0. The transition of the RDF flag from 0 to 1 can also be identified by an RXI interrupt. [3] Serial reception continuation procedure: RDF = 1? Yes Read receive data in SCFRDR, and clear RDF flag in SCFSR to 0 No All data received? Yes Clear RE bit in SCSCR to 0 End of reception [3] To continue serial reception, read at least the receive trigger set number of receive data bytes from SCFRDR, read 1 from the RDF flag, then clear the RDF flag to 0. The number of receive data bytes in SCFRDR can be ascertained by reading SCFRDR.
Start of reception Read ORER flag in SCLSR
ORER = 1?
Figure 14.15 Sample Flowchart for Receiving Serial Data (1)
Error handling No
ORER = 1? Yes Overrun error handling
Clear ORER flag in SCLSR to 0
End
Figure 14.16 Sample Flowchart for Receiving Serial Data (2)
Rev. 6.00 Jun. 12, 2007 Page 380 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
In receiving, the SCIF operates as follows: 1. The SCIF synchronizes with serial clock input or output and initializes internally. 2. Receive data is shifted into SCRSR in order from the LSB to the MSB. After receiving the data, the SCIF checks the receive data can be loaded from SCRSR into SCFRDR or not. If this check is passed, the SCIF stores the received data in SCFRDR. If the check is not passed (overrun error is detected), further reception is prevented. 3. After setting RDF to 1, if the receive-data-full interrupt enable bit (RIE) is set to 1 in SCSCR, the SCIF requests a receive-data-full interrupt (RXI). If the ORER bit is set to 1 and the RIE bit or REIE bit in SCSCR is also set to 1, the SCIF requests a break interrupt (BRI). Figure 14.17 shows an example of SCIF receive operation.
Synchronization clock Serial data Bit 7 LSB Bit 0 MSB Bit 7 Bit 0 Bit 1 Bit 6 Bit 7
RDF
ORER
Data read from RXI RXI SCFRDR and interrupt interrupt request RDF flag cleared request to 0 by RXI interrupt handler BRI interrupt request by overrun error
One frame
Figure 14.17 Example of SCIF Receive Operation
Rev. 6.00 Jun. 12, 2007 Page 381 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Transmitting and Receiving Serial Data Simultaneously (Synchronous Mode): Figure 14.18 shows a sample flowchart for transmitting and receiving serial data simultaneously.
Initialization
[1] SCIF status check and transmit data write: Read SCFSR and check that the TDFE flag is set to 1, then write transmit data to SCFTDR. Read the TDFE and TEND flags while they are 1, then clear them to 0. The transition of the TDFE flag from 0 to 1 can also be identified by a TXI interrupt. [2] Receive error handling: Read the ORER flag in SCLSR to identify any error, perform the appropriate error handling, then clear the ORER flag to 0. Reception cannot be resumed while the ORER flag is set to 1. [3] SCIF status check and receive data read:
Start of transmission and reception
Read TDFE flag in SCFSR
No
TDFE = 1? Yes Write transmit data to SCFTDR Read TDFE and TEND flags in SCFSR while they are 1, then clear them to 0
[1]
Read ORER flag in SCLSR Yes [2] No Read RDF flag in SCFSR Error handling
ORER = 1?
Read SCFSR and check that RDF = 1, then read the receive data in SCFRDR, and clear the RDF flag to 0. The transition of the RDF flag from 0 to 1 can also be identified by an RXI interrupt. [4] Serial transmission and reception continuation procedure: To continue serial transmission and reception, read 1 from the RDF flag and the receive data in SCFRDR, and clear the RDF flag to 0 before receiving the MSB in the current frame. Similarly, read 1 from the TDFE flag to confirm that writing is possible before transmitting the MSB in the current frame. Then write data to SCFTDR and clear the TDFE flag to 0.
No
RDF = 1? Yes Read receive data in SCFRDR, and clear RDF flag in SCFSR to 0
[3]
No
All data received? Yes Clear TE and RE bits in SCSCR to 0 [4] Note: When switching from a transmit operation or receive operation to simultaneous transmission and reception operations, clear the TE and RE bits to 0, and then set them simultaneously to 1.
End of transmission and reception
Figure 14.18 Sample Flowchart for Transmitting/Receiving Serial Data
Rev. 6.00 Jun. 12, 2007 Page 382 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
14.5
SCIF Interrupts
The SCIF has four interrupt sources: transmit-FIFO-data-empty (TXI), receive-error (ERI), receive-data-full (RXI), and break (BRI). Table 14.11 shows the interrupt sources and their order of priority. The interrupt sources are enabled or disabled by means of the TIE, RIE, and REIE bits in SCSCR. A separate interrupt request is sent to the interrupt controller for each of these interrupt sources. When TXI request is enabled by TIE bit and the TDFE flag in the serial status register (SCFSR) is set to 1, a TXI interrupt request is generated. When RXI request is enabled by RIE bit and the RDF or DR flag in SCFSR is set to 1, an RXI interrupt request is generated. The RXI interrupt request caused by DR flag is generated only in asynchronous mode. When BRI request is enabled by RIE bit or REIE bit and the BRK flag in SCFSR or ORER flag in SCLSR is set to 1, a BRI interrupt request is generated. When ERI request is enabled by RIE bit or REIE bit and the ER flag in SCFCR is set to 1, an ERI interrupt request is generated. When the RIE bit is set to 0 and the REIE bit is set to 1, SCIF request ERI interrupt and BRI interrupt without requesting RXI interrupt. The TXI interrupt indicates that transmit data can be written, and the RXI interrupt indicates that there is receive data in SCFRDR. Table 14.11 SCIF Interrupt Sources
Interrupt Source ERI RXI BRI TXI Description Interrupt initiated by receive error (ER) Interrupt Enable Bit RIE or REIE Priority on Reset Release High
Interrupt initiated by receive data FIFO full (RDF) or RIE data ready (DR) Interrupt initiated by break (BRK) or overrun error (ORER) Interrupt initiated by transmit FIFO data empty (TDFE) RIE or REIE TIE Low
Rev. 6.00 Jun. 12, 2007 Page 383 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
14.6
Serial Port Register (SCSPTR) and SCIF Pins
The relationship between SCSPTR and the SCIF pins is shown in figures 14.19 to 14.23.
Reset R Q D RTSIO C SPTRW Reset RTS R Q D RTSDT C SPTRW Modem control enable signal* RTS signal Bit 6 Bit 7
Internal data bus
SPTRR SPTRW: SPTRR: Note: SCSPTR write SCSPTR read
* The modem control function is specified for the RTS pin by setting the MCE bit in SCFCR.
Figure 14.19 RTSIO Bit, RTSDT bit, and RTS Pin
Rev. 6.00 Jun. 12, 2007 Page 384 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Reset R Q D CTSIO C SPTRW Reset R Q D CTSDT C SPTRW Bit 4 Bit 5
Internal data bus
CTS
Modem control enable signal* CTS signal
SPTRR SPTRW: SPTRR: Note: SCSPTR write SCSPTR read
* The modem control function is specified for the CTS pin by setting the MCE bit in SCFCR.
Figure 14.20 CTSIO Bit, CTSDT bit, and CTS Pin
Rev. 6.00 Jun. 12, 2007 Page 385 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Reset R Q D SCKIO C SPTRW Reset SCK2 R Q D SCKDT C SPTRW Clock output enable signal* Sirial clock output signal* Serial clock input signal* Serial input enable signal* Bit 2 Internal data bus Bit 3
SPTRR SPTRW: SPTRR: Note: SCSPTR write SCSPTR read
* These signals control the SCK pin according to the settings of the C/A bit in SCSMR and bits CKE1 and CKE0 in SCSCR.
Figure 14.21 SCKIO Bit, SCKDT bit, and SCK Pin
Reset R Q D SPBIO C SPTRW Reset TxD R Q D SPBDT C SPTRW Transmit enable signal Serial transmit data Bit 0 Bit 1
Internal data bus
SPTRW:
SCSPTR write
Figure 14.22 SPBIO Bit, SPBDT bit, and TxD Pin
Rev. 6.00 Jun. 12, 2007 Page 386 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
RxD Serial receive data
Internal data bus
Bit 0 SPTRR SPTRR: SCSPTR read
Figure 14.23 SPBDT bit and RxD Pin
Rev. 6.00 Jun. 12, 2007 Page 387 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
14.7
Usage Notes
Note the following when using the SCIF. 1. SCFTDR Writing and TDFE Flag The TDFE flag in the serial status register (SCFSR) is set when the number of transmit data bytes written in the transmit FIFO data register (SCFTDR) has fallen below the transmit trigger number set by bits TTRG1 and TTRG0 in the FIFO control register (SCFCR). After TDFE is set, transmit data up to the number of empty bytes in SCFTDR can be written, allowing efficient continuous transmission. However, if the number of data bytes written in SCFTDR is equal to or less than the transmit trigger number, the TDFE flag will be set to 1 again after being read as 1 and cleared to 0. TDFE clearing should therefore be carried out when SCFTDR contains more than the transmit trigger number of transmit data bytes. The number of transmit data bytes in SCFTDR can be found from the upper 8 bits of the FIFO data count register (SCFDR). 2. SCFRDR Reading and RDF Flag The RDF flag in the serial status register (SCFSR) is set when the number of receive data bytes in the receive FIFO data register (SCFRDR) has become equal to or greater than the receive trigger number set by bits RTRG1 and RTRG0 in the FIFO control register (SCFCR). After RDF is set, receive data equivalent to the trigger number can be read from SCFRDR, allowing efficient continuous reception. However, if the number of data bytes in SCFRDR is equal to or greater than the trigger number, the RDF flag will be set to 1 again if it is cleared to 0. RDF should therefore be cleared to 0 after being read as 1 after all the receive data has been read. The number of receive data bytes in SCFRDR can be found from the lower 8 bits of the FIFO data count register (SCFDR). 3. Break Detection and Processing Break signals can be detected by reading the RxD pin directly when a framing error (FER) is detected. In the break state the input from the RxD pin consists of all 0s, so the FER flag is set and the parity error flag (PER) may also be set. Note that, although transfer of receive data to SCFRDR is halted in the break state, the SCIF receiver continues to operate.
Rev. 6.00 Jun. 12, 2007 Page 388 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
4. Sending a Break Signal The I/O condition and level of the TxD pin are determined by the SPBIO and SPBDT bits in the serial port register (SCSPTR). This feature can be used to send a break signal. Until TE bit is set to 1 (enabling transmission) after initializing, TxD pin does not work. During the period, mark status is performed by SPBDT bit. Therefore, the SPBIO and SPBDT bits should be set to 1 (high level output). To send a break signal during serial transmission, clear the SPBDT bit to 0 (designating low level), then clear the TE bit to 0 (halting transmission). When the TE bit is cleared to 0, the transmitter is initialized regardless of the current transmission state, and 0 is output from the TxD pin. 5. Receive Data Sampling Timing and Receive Margin (Asynchronous Mode) The SCIF operates on a base clock with a frequency of 16 times the transfer rate. In reception, the SCIF synchronizes internally with the fall of the start bit, which it samples on the base clock. Receive data is latched at the rising edge of the eighth base clock pulse. The timing is shown in figure 14.24.
16 clocks 8 clocks
0 1 2 3 4 5 6 7 8 9 10 1112 1314 15 0 1 2 3 4 5 6 7 8 9 10 1112 1314 15 0 1 2 3 4 5
Base clock -7.5 clocks +7.5 clocks
Receive data (RxD) Synchronization sampling timing Data sampling timing
Start bit
D0
D1
Figure 14.24 Receive Data Sampling Timing in Asynchronous Mode The receive margin in asynchronous mode can therefore be expressed as shown in equation 1.
Rev. 6.00 Jun. 12, 2007 Page 389 of 610 REJ09B0131-0600
Section 14 Serial Communication Interface with FIFO (SCIF)
Equation 1:
M = (0.5 D - 0.5 1 ) = (L - 0.5) F (1+F) x 100 % 2N N
Where: M: Receive margin (%) N: Ratio of clock frequency to bit rate (N = 16) D: Clock duty cycle (D = 0 to 1.0) L: Frame length (L = 9 to 12) F: Absolute deviation of clock frequency From equation 1, if F = 0 and D = 0.5, the receive margin is 46.875%, as given by equation 2. Equation 2: When D = 0.5 and F = 0: M = (0.5 - 1/(2 x 16)) x 100% = 46.875% This is a theoretical value. A reasonable margin to allow in system designs is 20% to 30%. 6. Prohibited Multiple Pin Allocation for Channel 1 Although signal SCK1, RxD1, or TxD1 can be assigned to pin PD4 or PE20, either of the pin must be selected. For example, if signal SCK1 is assigned to both pins PD4 and PE20, correct operation of the SCIF is not guaranteed. Similarly, signal SCK1, RxD1, or TxD1 can be assigned to pin PD3 or PE19 and pin PD2 or PE18, respectively. However if these signals are assigned to both corresponding pins, correct operation of the SCIF is not guaranteed. 7. States of the TxD and RTS Pins When the TE Bit is Cleared The TxDi (i = 0, 1, 2) and RTSj (j = 0, 1) pins usually function as output pins during serial communication. However, even if these functions are selected by the pin function controller (PFC), these pins are in the high impedance state as long as the TE bit in SCSCRi (i = 0, 1, 2) is cleared. To make these pins always function as output pins (regardless of the value of the TE bit), set SCPTRi (i = 0, 1, 2) and PFC in the following order. a. Set the SPBIO and SPBDT bits in SCPTRi (i = 0, 1, 2). Set the RTSIO and RTSDT bits in SCPTRj (j = 0, 1). b. Select the TxDi (i = 0, 1, 2) and RTSj (j = 0, 1) pins by the PFC. 8. Interval from when the TE bit in SCSCR is Set to 1 until a Start Bit is Transmitted in Asynchronous Mode In the SCIF included in former products, a start bit is transmitted after the internal equivalent to one frame. In the SCIF included in this product, however, a start bit is transmitted directly after the TE bit is set to 1.
Rev. 6.00 Jun. 12, 2007 Page 390 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
Section 15 Host Interface (HIF)
This LSI incorporates a host interface (HIF) for use in high-speed transfer of data between external devices which cannot utilize the system bus. The HIF allows external devices to read from and write to 2 kbytes (1 kbyte x 2 banks) of the onchip RAM exclusively for HIF use (HIFRAM) within this LSI, in 32-bit units. Interrupts issued to this LSI by an external device, interrupts sent from this LSI to the external device, and DMA transfer requests sent from this LSI to the external device are also supported. By using HIFRAM and these interrupt functions, software-based data transfer between external devices and this LSI becomes possible, and connection to external devices not releasing bus mastership is enabled. Using HIFRAM, the HIF also supports HIF boot mode allowing this LSI to be booted.
15.1
Features
The HIF has the following features. * An external device can read from or write to HIFRAM in 32-bit units via the HIF pins (access in 8-bit or 16-bit units not allowed). The on-chip CPU can read from or write to HIFRAM in 8bit, 16-bit, or 32-bit units, via the internal peripheral bus. The HIFRAM access mode can be specified as bank mode or non-bank mode. * When an external device accesses HIFRAM via the HIF pins, automatic increment of addresses and the endian can be specified with the HIF internal registers. * By writing to specific bits in the HIF internal registers from an external device, or by accessing the end address of HIFRAM from the external device, interrupts (internal interrupts) can be issued to the on-chip CPU. Conversely, by writing to specific bits in the HIF internal registers from the on-chip CPU, interrupts (external interrupts) or DMAC transfer requests can be sent from the on-chip CPU to the external device. * There are seven interrupt source bits each for internal interrupts and external interrupts. Accordingly, software control of 128 different interrupts is possible, enabling high-speed data transfer using interrupts. * In HIF boot mode, this LSI can be booted from HIFRAM by an external device storing the instruction code in HIFRAM.
IFHIF00A_000020030900
Rev. 6.00 Jun. 12, 2007 Page 391 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
Figure 15.1 shows a block diagram of the HIF.
HIF
HIFDATA HIFSCR
HIFRAM HIFRAM
HIFD15 to HIFD00
HIFEICR HIFBCR HIFADR
HIFMCR
HIFIICR
Select
HIFCS HIFRS HIFWR HIFRD HIFMD HIFINT HIFDREQ HIFRDY HIFEBL Control circuit HIFI HIFBI
[Legend] HIFIDX: HIFGSR: HIFSCR: HIFMCR: HIFIICR: HIFEICR:
HIF index register HIF general status register HIF status/control register HIF memory control register HIF internal interrupt control register HIF external interrupt control register
HIFBICR
HIFGSR
HIFDTR
HIFIDX
HIFADR: HIFDATA: HIFBCR: HIFDTR: HIFBICR: HIFI: HIFIB:
HIF address register HIF data register HIF boot control register HIFDREQ trigger register HIF bank interrupt control register HIF interrupt (internal interrupt) HIF bank interrupt (internal interrupt)
Figure 15.1 Block Diagram of HIF
Rev. 6.00 Jun. 12, 2007 Page 392 of 610 REJ09B0131-0600
Internal bus
Section 15 Host Interface (HIF)
15.2
Input/Output Pins
Table 15.1 shows the HIF pin configuration. Table 15.1 Pin Configuration
Name HIF data pins HIF chip select HIF register select Abbreviation HIFD15 to HIFD00 HIFCS HIFRS I/O I/O Input Input Description Address, data, or command input/output to the HIF Chip select input to the HIF Switching between HIF access types 0: Normal access (other than below) 1: Index register write or status register read HIF write HIF read HIF interrupt HIF mode HIFWR HIFRD HIFINT HIFMD Input Input Write strobe signal. Low level is input when an external device writes data to the HIF. Read strobe signal. Low level is input when an external device reads data from the HIF.
Output Interrupt request to an external device from the HIF Input Selects whether or not this LSI is started up in HIF boot mode. If a power-on reset is canceled when high level is input, this LSI is started up in HIF boot mode.
HIFDMAC transfer request HIF boot ready
HIFDREQ HIFRDY
Output To an external device, DMAC transfer request with HIFRAM as the destination Output Indicates that the HIF reset is canceled in this LSI and access from an external device to the HIF can be accepted. After 10 clock cycles (max.) of the peripheral clock following negate of the reset input pin of this LSI, this pin is asserted.
HIF pin enable
HIFEBL
Input
All HIF pins other than this pin are asserted by high-level input.
Rev. 6.00 Jun. 12, 2007 Page 393 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
15.3
15.3.1
Parallel Access
Operation
The HIF can be accessed by combining the HIFCS, HIFRS, HIFWR, and HIFRD pins. Table 15.2 shows the correspondence between combinations of these signals and HIF operations. Table 15.2 HIF Operations
HIFCS 1 0 0 0 0 0 0 [Legend] x: Don't care HIFRS x 0 0 1 1 x x HIFWR x 1 0 1 0 1 0 HIFRD x 0 1 0 1 1 0 Operation No operation (NOP) Read from register specified by HIFIDX[7:0] Write to register specified by HIFIDX[7:0] Read from status register (HIFGSR[7:0]) Write to index register (HIFIDX[7:0]) No operation (NOP) Setting prohibited
15.3.2
Connection Method
When connecting the HIF to an external device, a method like that shown in figure 15.2 should be used.
External device CS A02 WR RD D15 to D00 HIFCS HIFRS HIFWR HIFRD HIFD15 to HIFD00 HIF
Figure 15.2 HIF Connection Example
Rev. 6.00 Jun. 12, 2007 Page 394 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
15.4
Register Descriptions
The HIF has the following registers. * * * * * * * * * * * HIF index register (HIFIDX) HIF general status register (HIFGSR) HIF status/control register (HIFSCR) HIF memory control register (HIFMCR) HIF internal interrupt control register (HIFIICR) HIF external interrupt control register (HIFEICR) HIF address register (HIFADR) HIF data register (HIFDATA) HIF boot control register (HIFBCR) HIFDREQ trigger register (HIFDTR) HIF bank interrupt control register (HIFBICR) HIF Index Register (HIFIDX)
15.4.1
HIFIDX is a 32-bit register used to specify the register read from or written to by an external device when the HIFRS pin is held low. HIFIDX can be only read by the on-chip CPU. HIFIDX can be only written to by an external device while the HIFRS pin is driven high.
Bit Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0.
31 to 8 --
Rev. 6.00 Jun. 12, 2007 Page 395 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
Bit 7 6 5 4 3 2
Bit Name REG5 REG4 REG3 REG2 REG1 REG0
Initial Value 0 0 0 0 0 0
R/W R/W* R/W* R/W* R/W* R/W* R/W*
Description HIF Internal Register Select These bits specify which register among HIFGSR, HIFSCR, HIFMCR, HIFIICR, HIFEICR, HIFADR, HIFDATA, and HIFBCR is accessed by an external device. 000000: HIFGSR 000001: HIFSCR 000010: HIFMCR 000011: HIFIICR 000100: HIFEICR 000101: HIFADR 000110: HIFDATA 001111: HIFBCR Other than above: Setting prohibited
Rev. 6.00 Jun. 12, 2007 Page 396 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
Bit 1 0
Bit Name BYTE1 BYTE0
Initial Value 0 0
R/W R/W* R/W*
Description Internal Register Byte Specification These bits specify in advance the target word location before the external device accesses a register among HIFGSR, HIFSCR, HIFMCR, HIFIICR, HIFEICR, HIFADR, HIFDATA, and HIFBCR. * When HIFSCR.BO = 0 00: Bits 31 to 16 in register 01: Setting prohibited 10: Bits 15 to 0 in register 11: Setting prohibited * When HIFSCR.BO = 1 00: Bits 15 to 0 in register 01: Setting prohibited 10: Bits 31 to 16 in register 11: Setting prohibited However, when HIFDATA is selected using bits REG5 to REG0, each time reading or writing of HIFDATA occurs, these bits change according to the following rule. 00 10 00 10... repeated
Note:
*
This bit can be only written to by an external device while the HIFRS pin is held high. It cannot be written to by the on-chip CPU.
Rev. 6.00 Jun. 12, 2007 Page 397 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
15.4.2
HIF General Status Register (HIFGSR)
HIFGSR is a 32-bit register, which can be freely used for handshaking between an external device connected to the HIF and the software of this LSI. HIFGSR can be read from and written to by the on-chip CPU. Reading from HIFGSR by an external device should be performed with the HIFRS pin high, or HIFGSR specified by bits REG5 to REG0 in HIFIDX and the HIFRS pin low. Writing to HIFGSR by an external device should be performed with HIFGSR specified by bits REG5 to REG0 in HIFIDX and the HIFRS pin low.
Bit 31 to 16 Bit Name -- Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 15 to 0 STATUS15 to All 0 STATUS0 R/W General Status This register can be read from and written to by an external device connected to the HIF, and by the on-chip CPU. These bits are initialized only at a power-on reset.
15.4.3
HIF Status/Control Register (HIFSCR)
HIFSCR is a 32-bit register used to control the HIFRAM access mode and endian setting. HIFSCR can be read from and written to by the on-chip CPU. Access to HIFSCR by an external device should be performed with HIFSCR specified by bits REG5 to REG0 in HIFIDX and the HIFRS pin low.
Bit 31 to 12 Bit Name -- Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0.
Rev. 6.00 Jun. 12, 2007 Page 398 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
Bit 11 10
Bit Name DMD DPOL
Initial Value 0 0
R/W R/W R/W
Description DREQ Mode DREQ Polarity Controls the assert mode for the HIFDREQ pin. For details on the negate timing, see section 15.8, External DMAC Interface. 00: For a DMAC transfer request to an external device, low level is generated at the HIFDREQ pin. The default for the HIFDREQ pin is high-level output. 01: For a DMAC transfer request to an external device, high level is generated at the HIFDREQ pin. The default for the HIFDREQ pin is low-level output. 10: For a DMAC transfer request to an external device, falling edge is generated at the HIFDREQ pin. The default for the HIFDREQ pin is high-level output. 11: For a DMAC transfer request to an external device, rising edge is generated at the HIFDREQ pin. The default for the HIFDREQ pin is low-level output.
9 8
BMD BSEL
0 0
R/W R/W
HIFRAM Bank Mode HIFRAM Bank Select Controls the HIFRAM access mode. 00: Both an external device and the on-chip CPU can access bank 0. When access by both of these conflict, even though the access addresses differ, access by the external device is processed before access by the on-chip CPU. Bank 1 cannot be accessed. 01: Both an external device and the on-chip CPU can access bank 1. When access by both of these conflict, even though the access addresses differ, access by the external device is processed before access by the on-chip CPU. Bank 0 cannot be accessed. 10: An external device can access only bank 0 while the on-chip CPU can access only bank 1. 11: An external device can access only bank 1 while the on-chip CPU can access only bank 0.
Rev. 6.00 Jun. 12, 2007 Page 399 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
Bit 7
Bit Name --
Initial Value 0
R/W R
Description Reserved This bit is always read as 0. The write value should always be 0.
6
--
1
R
Reserved This bit is always read as 1. The write value should always be 1.
5
MD1
0/1
R
HIF Mode 1 Indicates whether this LSI was started up in HIF boot mode or non-HIF boot mode. This bit stores the value of the HIFMD pin sampled at a power-on reset 0: Started up in non-HIF boot mode (booted from the memory connected to area 0) 1: Started up in HIF boot mode (booted from HIFRAM)
4 to 2
--
All 0
R
Reserved These bits are always read as 0. The write value should always be 0.
1
EDN
0
R/W
Endian for HIFRAM Access Specifies the byte order when HIFRAM is accessed by the on-chip CPU. 0: Big endian (MSB first) 1: Little endian (LSB first)
0
BO
0
R/W
Byte Order for Access of All HIF Registers Including HIFDATA Specifies the byte order when an external device accesses all HIF registers including HIFDATA. 0: Big endian (MSB first) 1: Little endian (LSB first)
Rev. 6.00 Jun. 12, 2007 Page 400 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
15.4.4
HIF Memory Control Register (HIFMCR)
HIFMCR is a 32-bit register used to control HIFRAM. HIFMCR can be only read by the on-chip CPU. Access to HIFMCR by an external device should be performed with HIFMCR specified by bits REG5 to REG0 in HIFIDX and the HIFRS pin low.
Bit Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 7 LOCK 0 R/W* Lock This bit is used to lock the access direction (read or write) for consecutive access of HIFRAM by an external device via HIFDATA. When this bit is set to 1, the values of the RD and WT bits set at the same time are held until this bit is next cleared to 0. When the RD bit and this bit are simultaneously set to 1, consecutive read mode is entered. When the WT bit and this bit are simultaneously set to 1, consecutive write mode is entered. Both the RD and WT bits should not be set to 1 simultaneously. 6 -- 0 R Reserved This bit is always read as 0. The write value should always be 0. 5 WT 0 R/W* Write When this bit is set to 1, the HIFDATA value is written to the HIFRAM position corresponding to HIFADR. If this bit and the LOCK bit are set to 1 simultaneously, HIFRAM consecutive write mode is entered, and highspeed data transfer becomes possible. This mode is maintained until this bit is next cleared to 0, or until the LOCK bit is cleared to 0. If the LOCK bit is not simultaneously set to 1 with this bit, writing to HIFRAM is performed only once. Thereafter, the value of this bit is automatically cleared to 0. 4 -- 0 R Reserved This bit is always read as 0. The write value should always be 0.
31 to 8 --
Rev. 6.00 Jun. 12, 2007 Page 401 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
Bit 3
Bit Name RD
Initial Value 0
R/W R/W*
Description Read When this bit is set to 1, the HIFRAM data corresponding to HIFADR is fetched to HIFDATA. If this bit and the LOCK bit are set to 1 simultaneously, HIFRAM consecutive read mode is entered, and highspeed data transfer becomes possible. This mode is maintained until this bit is next cleared to 0, or until the LOCK bit is cleared to 0. If the LOCK bit is not simultaneously set to 1 with this bit, reading of HIFRAM is performed only once. Thereafter, the value of this bit is automatically cleared to 0.
2, 1
--
All 0
R
Reserved These bits are always read as 0. The write value should always be 0.
0
AI/AD
0
R/W*
Address Auto-Increment/Decrement This bit is valid only when the LOCK bit is 1. The value of HIFADR is automatically incremented by 4 or decremented by 4 according to the setting of this bit each time reading or writing of HIFRAM is performed. 0: Auto-increment mode (+4) 1: Auto-decrement mode (-4)
Note:
*
This bit can be only written to by an external device when the HIFRS pin is low. It cannot be written to by the on-chip CPU. Changing the HIFRAM banks accessible from an external device by setting the BMD and BSEL bits in HIFSCR does not affect the setting of this bit.
Rev. 6.00 Jun. 12, 2007 Page 402 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
15.4.5
HIF Internal Interrupt Control Register (HIFIICR)
HIFIICR is a 32-bit register used to issue interrupts from an external device connected to the HIF to the on-chip CPU. Access to HIFIICR by an external device should be performed with HIFIICR specified by bits REG5 to REG0 in HIFIDX and the HIFRS pin low.
Bit Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 7 6 5 4 3 2 1 0 IIC6 IIC5 IIC4 IIC3 IIC2 IIC1 IIC0 IIR 0 0 0 0 0 0 0 0 R/W R/W R/W R/W R/W R/W R/W R/W Internal Interrupt Request While this bit is 1, an interrupt request (HIFI) is issued to the on-chip CPU. Internal Interrupt Source These bits specify the source for interrupts generated by the IIR bit. These bits can be written to from both an external device and the on-chip CPU. By using these bits, fast execution of interrupt exception handling is possible. These bits are completely under software control, and their values have no effect on the operation of this LSI.
31 to 8 --
15.4.6
HIF External Interrupt Control Register (HIFEICR)
HIFEICR is a 32-bit register used to issue interrupts to an external device connected to the HIF from this LSI. Access to HIFEICR by an external device should be performed with HIFEICR specified by bits REG5 to REG0 in HIFIDX and the HIFRS pin low.
Bit Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0.
31 to 8 --
Rev. 6.00 Jun. 12, 2007 Page 403 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
Bit 7 6 5 4 3 2 1 0
Bit Name EIC6 EIC5 EIC4 EIC3 EIC2 EIC1 EIC0 EIR
Initial Value 0 0 0 0 0 0 0 0
R/W R/W R/W R/W R/W R/W R/W R/W R/W
Description External Interrupt Source These bits specify the source for interrupts generated by the EIR bit. These bits can be written to from both an external device and the on-chip CPU. By using these bits, fast execution of interrupt exception handling is possible. These bits are completely under software control, and their values have no effect on the operation of this LSI. External Interrupt Request While this bit is 1, the HIFINT pin is asserted to issue an interrupt request to an external device from this LSI.
15.4.7
HIF Address Register (HIFADR)
HIFADR is a 32-bit register which indicates the address in HIFRAM to be accessed by an external device. When using the LOCK bit setting in HIFMCR to specify consecutive access of HIFRAM, auto-increment (+4) or auto-decrement (-4) of the address, according to the AI/AD bit setting in HIFMCR, is performed automatically, and HIFADR is updated. HIFADR can be only read by the on-chip CPU. Access to HIFADR by an external device should be performed with HIFADR specified by bits REG5 to REG0 in HIFIDX and the HIFRS pin low.
Bit 31 to 10 Bit Name -- Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 9 to 2 A9 to A2 All 0 R/W* HIFRAM Address Specification These bits specify the address of HIFRAM to be accessed by an external device, with 32-bit boundary. 1, 0 -- All 0 R Reserved These bits are always read as 0. The write value should always be 0. Note: * This bit can be only written to by an external device when the HIFRS pin is low. It cannot be written to by the on-chip CPU.
Rev. 6.00 Jun. 12, 2007 Page 404 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
15.4.8
HIF Data Register (HIFDATA)
HIFDATA is a 32-bit register used to hold data to be written to HIFRAM and data read from HIFRAM for external device accesses. If HIFDATA is not used when accessing HIFRAM, it can be used for data transfer between an external device connected to the HIF and the on-chip CPU. HIFDATA can be read from and written to by the on-chip CPU. Access to HIFDATA by an external device should be performed with HIFDATA specified by bits REG5 to REG0 in HIFIDX and the HIFRS pin low.
Bit 31 to 0 Bit Name D31 to D0 Initial Value All 0 R/W R/W Description 32-bit Data
15.4.9
HIF Boot Control Register (HIFBCR)
HIFBCR is a 32-bit register for exclusive control of an external device and the on-chip CPU regarding access of HIFRAM. HIFBCR can be only read by the on-chip CPU. Access to HIFBCR by an external device should be performed with HIFBCR specified by bits REG5 to REG0 in HIFIDX and the HIFRS pin low.
Bit 31 to 8 Bit Name -- Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 7 to 1 -- All 0 R/W AC-Bit Writing Assistance These bits should be used to write the bit pattern (H'A5) needed to set the AC bit to 1. These bits are always read as 0.
Rev. 6.00 Jun. 12, 2007 Page 405 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
Bit 0
Bit Name AC
Initial Value 0/1
R/W R/W
Description HIFRAM Access Exclusive Control Controls accessing of HIFRAM by the on-chip CPU for the HIFRAM bank selected by the BMD and BSEL bits in HIFSCR as the bank allowed to be accessed by this LSI. 0: The on-chip CPU can perform reading/writing of HIFRAM. 1: When an HIFRAM read/write operation by the onchip CPU occurs, the CPU enters the wait state, and execution of the instruction is halted until this bit is cleared to 0. When booted in non-HIF boot mode, the initial value of this bit is 0. When booted in HIF boot mode, the initial value of this bit is 1. After an external device writes a boot program to HIFRAM via the HIF, clearing this bit to 0 boots the on-chip CPU from HIFRAM. When 1 is written to this bit by an external device, H'A5 should be written to bits 7 to 0 to prevent erroneous writing.
15.4.10 HIFDREQ Trigger Register (HIFDTR) HIFDTR is a 32-bit register. Writing to HIFDTR by the on-chip CPU asserts the HIFDREQ pin. HIFDTR cannot be accessed by an external device.
Bit 31 to 1 Bit Name -- Initial Value All 0 R/W R*
1
Description Reserved These bits are always read as 0. The write value should always be 0.
Rev. 6.00 Jun. 12, 2007 Page 406 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
Bit 0
Bit Name DTRG
Initial Value 0
R/W R/W* *
1 2
Description HIFDREQ Trigger When 1 is written to this bit, the HIFDREQ pin is asserted according to the setting of the DMD and DPOL bits in HIFSCR. This bit is automatically cleared to 0 in synchronization with negate of the HIFDREQ pin. Though this bit can be set to 1 by the on-chip CPU, it cannot be cleared to 0. To avoid conflict between clearing of this bit by negate of the HIFDREQ pin and setting of this bit by the on-chip CPU, make sure this bit is cleared to 0 before setting this bit to 1 by the on-chip CPU.
Notes: 1. This bit cannot be accessed by an external device. It can be accessed only by the onchip CPU. 2. Writing 0 to this bit by the on-chip CPU is ignored.
15.4.11 HIF Bank Interrupt Control Register (HIFBICR) HIFBICR is a 32-bit register that controls HIF bank interrupts. HIFBICR cannot be accessed by an external device.
Bit 31 to 2 Bit Name -- Initial Value All 0 R/W R*
1
Description Reserved These bits are always read as 0. The write value should always be 0.
1
BIE
0
R/W*1
Bank Interrupt Enable Enables or disables a bank interrupt request (HIFBI) issued to the on-chip CPU. 0: HIFBI disabled 1: HIFBI enabled
Rev. 6.00 Jun. 12, 2007 Page 407 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
Bit 0
Bit Name BIF
Initial Value 0
R/W R/W* *
1 2
Description Bank Interrupt Request Flag While this bit is 1, a bank interrupt request (HIFBI) is issued to the on-chip CPU according to the setting of the BIE bit. In auto-increment mode (AI/AD bit in HIFMCR is 0), this bit is automatically set to 1 when an external device has completed access to the 32-bit data in the end address of HIFRAM and the HIFCS pin has been negated. In auto-decrement mode (AI/AD bit in HIFMCR is 1), this bit is automatically set to 1 when an external device has completed access to the 32-bit data in the start address of HIFRAM and the HIFCS pin has been negated. Though this bit can be cleared to 0 by the on-chip CPU, it cannot be set to 1. Make sure setting of this bit by HIFRAM access from an external device and clearing of this bit by the onchip CPU do not conflict using software.
Notes: 1. This bit cannot be accessed by an external device. It can only be accessed by the onchip CPU. 2. Writing 1 to this bit by the on-chip CPU is ignored.
Rev. 6.00 Jun. 12, 2007 Page 408 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
15.5
Memory Map
Table 15.3 shows the memory map of HIFRAM. Table 15.3 Memory Map
Classification Map from external device*
1
Start Address H'0000 H'F84E0000
End Address H'03FF H'F84E03FF
Memory Size 1 kbyte 1 kbyte
Map from on-chip CPU*1 *2
Notes: 1. Map for a single HIFRAM bank. Which bank is to be accessed by an external device or the on-chip CPU depends on the BMD and BSEL bits in HIFSCR. The mapping addresses are common between the banks. 2. Note that in HIF boot mode, bank 0 is selected, and the first 1 kbyte in each of the following address ranges are also mapped: H'00000000 to H'01FFFFFF (first-half 32 Mbytes of area 0 in the P0 area), H'20000000 to H'21FFFFFF (first-half 32 Mbytes of area 0 in the P0 area), H'40000000 to H'41FFFFFF (first-half 32 Mbytes of area 0 in the P0 area), H'60000000 to H'61FFFFFF (first-half 32 Mbytes of area 0 in the P0 area), H'80000000 to H'81FFFFFF (first-half 32 Mbytes of area 0 in the P1 area), H'A0000000 to H'A1FFFFFF (first-half 32 Mbytes of area 0 in the P2 area), and H'C0000000 to H'C1FFFFFF (first-half 32 Mbytes of area 0 in the P3 area). If an external device modifies HIFRAM when HIFRAM is accessed from the P0, P1, or P3 area with the cache enabled, coherency may not be ensured. When the cache is enabled, accessing HIFRAM from the P2 area is recommended. In HIF boot mode, among the first-half 32 Mbytes of each area 0, access to the areas to which HIFRAM is not mapped is inhibited. Even in HIF boot mode, the second-half 32 Mbytes of area 0, area 3, area 4, area 5B, area 5, area 6B, and area 6 are mapped to the external memory as normally.
Rev. 6.00 Jun. 12, 2007 Page 409 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
15.6
Interface (Basic)
Figure 15.3 shows the basic read/write sequence. HIF read is defined by the overlap period of the HIFRD low-level period and HIFCS low-level period, and HIF write is defined by the overlap period of the HIFWR low-level period and HIFCS low-level period. The HIFRS signal indicates whether this is normal access or index/status register access; low level indicates normal access and high level indicates index/status register access.
Write cycle HIFCS HIFRS HIFRD HIFWR HIFD15 to HIFD00 WT_D RD_D Read cycle
Figure 15.3 Basic Timing for HIF Interface
Rev. 6.00 Jun. 12, 2007 Page 410 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
15.7
15.7.1
Interface (Details)
HIFIDX Write/HIFGSR Read
Writing of HIFIDX and reading of HIFGSR are shown in figure 15.4.
HIFIDX write cycle HIFGSR read cycle
HIFCS HIFRS HIFRD HIFWR HIFD15 to HIFD00 WT_D RD_D
Figure 15.4 HIFIDX Write and HIFGSR Read 15.7.2 Reading/Writing of HIF Registers other than HIFIDX and HIFGSR
As shown in figure 15.5, in reading and writing of HIF internal registers other than HIFIDX and HIFGSR, first HIFRS is held high and HIFIDX is written to in order to select the register to be accessed and the byte location. Then HIFRS is held low, and reading or writing of the register selected by HIFIDX is performed.
Index write Register write Register read
HIFCS HIFRS HIFRD HIFWR HIFD15 to HIFD00
HIFIDX
WT_D
RD_D
Register selection
Figure 15.5 HIF Register Settings
Rev. 6.00 Jun. 12, 2007 Page 411 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
15.7.3
Consecutive Data Writing to HIFRAM by External Device
Figure 15.6 shows the timing chart for consecutive data transfer from an external device to HIFRAM. As shown in this timing chart, by setting the start address and the data to be written first, consecutive data transfer can subsequently be performed.
HIFCS HIFRS HIFRD HIFWR HIFD15 to HIFD00
0016 AHAL 0018 D0D1 001A D2D3 000A 00A0 0018 D4D5 D6D7 D8D9
High level
HIFADR setting [15:8] = AH [7:0] = AL
Data for first write operation set in HIFDATA [31:24] = D0, [23:16] = D1, [15:8] = D2, [7:0] = D3
HIFMCR setting HIFDATA Consecutive data writing Consecutive write selection Auto-increment
Figure 15.6 Consecutive Data Writing to HIFRAM 15.7.4 Consecutive Data Reading from HIFRAM to External Device
Figure 15.7 shows the timing chart for consecutive data reading from HIFRAM to an external device. As this timing chart indicates, by setting the start address, data can subsequently be read out consecutively.
Rev. 6.00 Jun. 12, 2007 Page 412 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
HIFCS HIFRS HIFRD HIFWR HIFD15 to HIFD00
0016 AHAL 000A 0088 0018 D0D1 D2D3 D4D5 D6D7 D8D9 DADB DCDD
HIFADR setting HIFMCR setting HIFDATA [15:8] = AH Consecutive read selection [7:0] = AL Auto-increment
Consecutive data reading
Figure 15.7 Consecutive Data Reading from HIFRAM
15.8
External DMAC Interface
Figures 15.8 to 15.11 show the HIFDREQ output timing. The start of the HIFDREQ assert synchronizes with the DTRG bit in HIFDTR being set to 1. The HIFDREQ negate timing and assert level are determined by the DMD and DPOL bits in HIFSCR, respectively. When the external DMAC is specified to detect low level of the HIFDREQ signal, set DMD = 0 and DPOL = 0. After writing 1 to the DTRG bit, the HIFDREQ signal remains low until low level is detected for both the HIFCS and HIFRS signals. In this case, when the HIFDREQ signal is used, make sure that the setup time (HIFCS assertion to HIFRS settling) and the hold time (HIFRS hold to HIFCS negate) are satisfied. If tHIFAS and tHIFAH stipulated in section 21.4.9, HIF Timing, are not satisfied, the HIFDREQ signal may be negated unintentionally.
DTRG bit DPOL bit Asserted in synchronization with the DTRG bit being set by the on-chip CPU. HIFDREQ The DTRG bit is cleared simultaneously with HIFDREQ negate.
Negated when HIFCS = HIFRS = low level. Latency is tPCYC (peripheral clock cycle) x 3 cyc or less. HIFCS HIFRS
Figure 15.8 HIFDREQ Timing (When DMD = 0 and DPOL = 0)
Rev. 6.00 Jun. 12, 2007 Page 413 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
When the external DMAC is specified to detect high level of the HIFDREQ signal, set DMD = 0 and DPOL = 1. At the time the DPOL bit is set to 1, HIFDREQ becomes low. Then after writing 1 to the DTRG bit, HIFDREQ remains high until low level is detected for both the HIFCS and HIFRS signals. In this case, when the HIFDREQ signal is used, make sure that the setup time (HIFCS assertion to HIFRS settling) and the hold time (HIFRS hold to HIFCS negate) are satisfied. If tHIFAS and tHIFAH stipulated in section 21.4.9, HIF Timing, are not satisfied, the HIFDREQ signal may be negated unintentionally.
DTRG bit DPOL bit
Negated in synchronization with the DPOL bit being set by the on-chip CPU. Asserted in synchronization with the DTRG bit being set by the on-chip CPU. The DTRG bit is cleared simultaneously with HIFDREQ negate.
HIFDREQ
Negated when HIFCS = HIFRS = low level. Latency is tPCYC (peripheral clock cycle) x 3 cyc or less.
HIFCS HIFRS
Figure 15.9 HIFDREQ Timing (When DMD = 0 and DPOL = 1) When the external DMAC is specified to detect the falling edge of the HIFDREQ signal, set DMD = 1 and DPOL = 0. After writing 1 to the DTRG bit, a low pulse of 32 peripheral clock cycles is generated at the HIFDREQ pin.
DTRG bit DPOL bit
Asserted in synchronization with the DTRG bit being set by the on-chip CPU. The DTRG bit is cleared simultaneously with HIFDREQ negate. After assert, negated when tPCYC (peripheral clock cycle) x 32 cyc have elapsed.
HIFDREQ
Figure 15.10 HIFDREQ Timing (When DMD = 1 and DPOL = 0)
Rev. 6.00 Jun. 12, 2007 Page 414 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
When the external DMAC is specified to detect the rising edge of the HIFDREQ signal, set DMD = 1 and DPOL = 1. At the time the DPOL bit is set to 1, HIFDREQ becomes low. Then after writing 1 to the DTRG bit, a low pulse of 32 peripheral clock cycles is generated at the HIFDREQ pin.
DTRG bit DPOL bit
Negated in synchronization with the DPOL bit being set by the on-chip CPU. Asserted in synchronization with the DTRG bit being set by the on-chip CPU. The DTRG bit is cleared simultaneously with HIFDREQ negate.
HIFDREQ
After assert, negated when tPCYC (peripheral clock cycle) x 32 cyc have elapsed.
Figure 15.11 HIFDREQ Timing (When DMD = 1 and DPOL = 1) When the external DMAC supports intermittent operating mode (block transfer mode), efficient data transfer can be implemented by using the HIFRAM consecutive access and bank functions.
Rev. 6.00 Jun. 12, 2007 Page 415 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
Table 15.4 Consecutive Write Procedure to HIFRAM by External DMAC
External Device No. 1 2 3 CPU HIF initial setting DMAC initial setting Set HIFADR to HIFRAM end address -8 Select HIFDATA and write dummy data (4 bytes) to HIFDATA Set HIFRAM consecutive write with address increment in HIFMCR Select HIFDATA and write dummy data (4 bytes) to HIFDATA HIF bank interrupt occurs HIFRAM bank switching by HIF bank interrupt handler (external device accesses bank 1 and onchip CPU accesses bank 0) Set DTRG bit to 1 DMAC HIF This LSI CPU HIF initial setting
4
5
6
7 8
Activate DMAC Consecutive data write to bank 1 in HIFRAM
Assert HIFDREQ
9
Write to end HIF bank address of bank interrupt 1 in HIFRAM occurs completes and operation halts Re-activate DMAC Consecutive data write to bank 0 in HIFRAM Assert HIFDREQ
HIFRAM bank switching by HIF bank interrupt handler (external device accesses bank 0 and onchip CPU accesses bank 1) Set DTRG bit to 1 Read data from bank 1 in HIFRAM
10 11
Rev. 6.00 Jun. 12, 2007 Page 416 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
External Device No. 12 CPU DMAC HIF Write to end HIF bank address of bank interrupt 0 in HIFRAM occurs completes and operation halts Re-activate DMAC Assert HIFDREQ
This LSI CPU HIFRAM bank switching by HIF bank interrupt handler (external device accesses bank 1 and onchip CPU accesses bank 0) Set DTRG bit to 1
13
Hereafter No. 11 to 13 are repeated. When a register other than HIFDATA is accessed (except that HIFGSR read with HIFRS = low), HIFRAM consecutive write is interrupted, and No. 3 to 6 need to be done again.
Table 15.5 Consecutive Read Procedure from HIFRAM by External DMAC
External Device No. 1 2 3 CPU HIF initial setting DMAC initial setting Set HIFADR to HIFRAM start address Set HIFRAM consecutive read with address increment in HIFMCR Select HIFDATA Write data to bank 1 in HIFRAM After writing data to end address of bank 1 in HIFRAM, perform HIFRAM bank switching (external device accesses bank 1 and onchip CPU accesses bank 0) Activate DMAC Assert HIFDREQ Set DTRG bit to 1 DMAC HIF This LSI CPU HIF initial setting
4
5 6 7
8
Rev. 6.00 Jun. 12, 2007 Page 417 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
External Device No. 9 CPU DMAC Consecutive data read from bank 1 in HIFRAM Read from end HIF bank address of bank interrupt 1 in HIFRAM occurs completes and operation halts Re-activate DMAC Consecutive data read from bank 0 in HIFRAM Read from end HIF bank address of bank interrupt 0 in HIFRAM occurs completes and operation halts Re-activate DMAC Assert HIFDREQ Assert HIFDREQ HIF
This LSI CPU Write data to bank 0 in HIFRAM
10
HIFRAM bank switching by HIF bank interrupt handler (external device accesses bank 0 and onchip CPU accesses bank 1) Set DTRG bit to 1 Write data to bank 1 in HIFRAM
11 12
13
HIFRAM bank switching by HIF bank interrupt handler (external device accesses bank 1 and onchip CPU accesses bank 0) Set DTRG bit to 1
14
Hereafter No. 12 to 14 are repeated. When a register other than HIFDATA is accessed (except that HIFGSR read with HIFRS = low), HIFRAM consecutive read is interrupted, and No. 3 to 5 need to be done again.
Rev. 6.00 Jun. 12, 2007 Page 418 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
15.9
Interface When External Device Power is Cut Off
When the power supply of an external device interfacing with the HIF is cut off, intermediate levels may be applied to the HIF input pins or the HIF output pins may drive an external device not powered, thus causing the device to be damaged. The HIFEBL pin is provided to prevent this from happening. The system power monitor block controls the HIFEBL pin in synchronization with the cutoff of the external device power so that all HIF pins can be set to the high-impedance state. Figure 15.12 shows an image of high-impedance control of the HIF pins. Tables 15.6 and 15.7 list the input/output control for the HIF pins.
HIFD15 to HIFD00 HIFCS HIFRS HIFWR HIFRD HIFMD
HIFINT HIFDREQ HIFRDY
HIFEBL
Figure 15.12 Image of High-Impedance Control of HIF Pins by HIFEBL Pin
Rev. 6.00 Jun. 12, 2007 Page 419 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
Table 15.6 Input/Output Control for HIF Pins
LSI Status HIFMD input level Reset State by RES Pin Low (Non-boot setting) The HIFEBL pin is a general input port and the HIF is not controlled by the signal input on this pin. General input port Reset Canceled by RES Pin High (After the reset canceled by boot setting) Low (After the reset canceled by non-boot setting)
High (Boot setting)
HIFEBL input level HIFRDY output control
Low Output buffer: On (Low output) Output buffer: Off
High Output buffer: On (Low output) Output buffer: Off Output buffer: Off I/O buffer: Off
Low Output buffer: Off
High Output buffer: On (Sequence output) Output buffer: On (Sequence output) Output buffer: On (Sequence output)
General input port at the 1 initial state * General input port at the 2 initial state*
HIFINT output control
General input port
Output buffer: Off Output buffer: Off I/O buffer: Off
General input port at the 2 initial state*
HIFDREQ Output buffer: output Off control HIFD 15 to HIFD0 I/O control I/O buffer: Off
General input port
General input port at the 2 initial state*
General input port
General input port at the I/O buffer 2 initial state* controlled according to states of HIFCS, HIFWR, and HIFRD
HIFCS input control HIFRS input control
Input buffer: Off Input buffer: Off
Input buffer: Off Input buffer: Off
General input port
Input buffer: Input buffer: General input port at the 2 Off On initial state* Input buffer: Input buffer: General input port at the 2 Off On initial state*
General input port
Rev. 6.00 Jun. 12, 2007 Page 420 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
LSI Status HIFMD input level
Reset State by RES Pin
Reset Canceled by RES Pin High (After the reset canceled by boot setting) Low (After the reset canceled by non-boot setting)
High (Boot setting)
Low (Non-boot setting) The HIFEBL pin is a general input port and the HIF is not controlled by the signal input on this pin. General input port
HIFEBL input level HIFWR input control HIFRD input control
Low Input buffer: Off Input buffer: Off
High Input buffer: Off Input buffer: Off
Low Input buffer: Off Input buffer: Off
High Input buffer: On Input buffer: On
General input port at the initial 1 state * General input port at the initial 2 state* General input port at the initial 2 state*
General input port
Notes: 1. The pin also functions as an HIFEBL pin by setting the PFC registers. 2. The pin also functions as an HIF pin by setting the PFC registers. When the HIF pin function is selected for the HIFEBL pin and this pin by setting the PFC registers, the input and/or output buffers are controlled according to the HIFEBL pin state. When the HIF pin function is not selected for the HIFEBL pin and is selected for this pin by setting the PFC registers, the input and/or output buffers are always turned off. This setting is prohibited.
Rev. 6.00 Jun. 12, 2007 Page 421 of 610 REJ09B0131-0600
Section 15 Host Interface (HIF)
Rev. 6.00 Jun. 12, 2007 Page 422 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
Section 16 Pin Function Controller (PFC)
The pin function controller (PFC) consists of registers that select multiplexed pin functions and input/output directions. Tables 16.1 to 16.5 show the multiplexed pins in this LSI. Table 16.6 shows the pin functions in each operating mode. Table 16.1 List of Multiplexed Pins (Port A)
Port A Function 1 (Related Module) PA16 input/output (port) PA17 input/output (port) PA18 input/output (port) PA19 input/output (port) PA20 input/output (port) PA21 input/output (port) PA22 input/output (port) PA23 input/output (port) PA24 input/output (port) PA25 input/output (port) Function 2 (Related Module) A16 output (BSC) A17 output (BSC) A18 output (BSC) A19 output (BSC) A20 output (BSC) A21 output (BSC) A22 output (BSC) A23 output (BSC) A24 output (BSC) A25 output (BSC) Function 3 (Related Module) Function 4 (Related Module)
Table 16.2 List of Multiplexed Pins (Port B)
Function 1 (Related Module) PB00 input/output (port) PB01 input/output (port) PB02 input/output (port) CKE output (BSC) Function 2 (Related Module) WAIT input (BSC) IOIS16 input (BSC) Function 3 (Related Module) Function 4 (Related Module)
Port B

Rev. 6.00 Jun. 12, 2007 Page 423 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
Port B
Function 1 (Related Module) PB03 input/output (port) PB04 input/output (port) PB05 input/output (port) PB06 input/output (port) PB07 input/output (port) PB08 input/output (port) PB09 input/output (port) PB10 input/output (port) PB11 input/output (port) PB12 input/output (port) PB13 input/output (port)
Function 2 (Related Module) CAS output (BSC) RAS output (BSC) ICIORD output (BSC) ICIOWR output (BSC) CE2B output (BSC) CS6B output (BSC) CE1B output (BSC) CE2A output (BSC) CS5B output (BSC) CS4 output (BSC) CS3 output (BSC) BS output (BSC) CE1A output (BSC)
Function 3 (Related Module)
Function 4 (Related Module)

Rev. 6.00 Jun. 12, 2007 Page 424 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
Table 16.3 List of Multiplexed Pins (Port C)
Port C Function 1 (Related Module) PC00 input/output (port) PC01 input/output (port) PC02 input/output (port) PC03 input/output (port) PC04 input/output (port) PC05 input/output (port) PC06 input/output (port) PC07 input/output (port) PC08 input/output (port) PC09 input/output (port) PC10 input/output (port) PC11 input/output (port) PC12 input/output (port) PC13 input/output (port) PC14 input/output (port) PC15 input/output (port) PC16 input/output (port) PC17 input/output (port) PC18 input/output (port) PC19 input/output (port) PC20 input/output (port) Function 2 (Related Module) MII_RXD0 input (EtherC) MII_RXD1 input (EtherC) MII_RXD2 input (EtherC) MII_RXD3 input (EtherC) MII_TXD0 output (EtherC) MII_TXD1 output (EtherC) MII_TXD2 output (EtherC) MII_TXD3 output (EtherC) RX_DV input (EtherC) RX_ER input (EtherC) RX_CLK input (EtherC) TX_ER output (EtherC) TX_EN output (EtherC) TX_CLK input (EtherC) COL input (EtherC) CRS input (EtherC) Function 3 (Related Module) Function 4 (Related Module)
MDIO input/output (EtherC) MDC output (EtherC) LNKSTA input (EtherC) EXOUT output (EtherC) WOL output (EtherC)
Table 16.4 List of Multiplexed Pins (Port D)
Port D Function 1 (Related Module) PD0 input/output (port) PD1 input/output (port) PD2 input/output (port) PD3 input/output (port) PD4 input/output (port) PD5 input/output (port) Function 2 (Related Module) IRQ0 input (INTC) IRQ1 input (INTC) IRQ2 input (INTC) IRQ3 input (INTC) IRQ4 input (INTC) IRQ5 input (INTC) Function 3 (Related Module) TxD1 output (SCIF) RxD1 input (SCIF) Function 4 (Related Module)
SCK1 input/output (SCIF) TxD2 output (SCIF)
Rev. 6.00 Jun. 12, 2007 Page 425 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
Port D
Function 1 (Related Module) PD6 input/output (port) PD7 input/output (port)
Function 2 (Related Module) IRQ6 input (INTC) IRQ7 input (INTC)
Function 3 (Related Module) RxD2 input (SCIF)
Function 4 (Related Module)
SCK2 input/output (SCIF)
Table 16.5 List of Multiplexed Pins (Port E)
Port
E
Function 1 (Related Module)
PE00 input/output (port) PE01 input/output (port) PE02 input/output (port) PE03 input/output (port) PE04 input/output (port) PE05 input/output (port) PE06 input/output (port) PE07 input/output (port) PE08 input/output (port) PE09 input/output (port) PE10 input/output (port) PE11 input/output (port) PE12 input/output (port) PE13 input/output (port) PE14 input/output (port) PE15 input/output (port) PE16 input/output (port) PE17 input/output (port) PE18 input/output (port) PE19 input/output (port) PE20 input/output (port) PE21 input/output (port) PE22 input/output (port) PE23 input/output (port) PE24 input/output (port)
Function 2 (Related Module)
HIFEBL input (HIF) HIFRDY output (HIF) HIFDREQ output (HIF) HIFMD input (HIF) HIFINT output (HIF) HIFRD input (HIF) HIFWR input (HIF) HIFRS input (HIF) HIFCS input (HIF)
Function 3 (Related Module)

Function 4 (Related Module)

HIFD00 input/output (HIF) HIFD01 input/output (HIF) HIFD02 input/output (HIF) HIFD03 input/output (HIF) HIFD04 input/output (HIF) HIFD05 input/output (HIF) HIFD06 input/output (HIF) TxD0 output (SCIF) HIFD07 input/output (HIF) RxD0 input (SCIF) HIFD08 input/output (HIF) SCK0 input/output (SCIF) HIFD09 input/output (HIF) TxD1 output (SCIF) HIFD10 input/output (HIF) RxD1 input (SCIF) HIFD11 input/output (HIF) SCK1 input/output (SCIF) HIFD12 input/output (HIF) RTS0 output (SCIF) HIFD13 input/output (HIF) CTS0 input (SCIF) HIFD14 input/output (HIF) RTS1 output (SCIF) HIFD15 input/output (HIF) CTS1 input (SCIF)
Rev. 6.00 Jun. 12, 2007 Page 426 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
Table 16.6 Pin Functions in Each Operating Mode
Not HIF Boot Mode Pin No. C14 B15 B14 C13 B13 C12 A13 B12 D11 A12 C11 D10 C10 A10 B10 D9 B6 C5 A5 B5 D5 C4 A3 D4 B3 A2 C8 D7 Initial Function A00 A01 A02 A03 A04 A05 A06 A07 A08 A09 A10 A11 A12 A13 A14 A15 PA16 PA17 PA18 PA19 PA20 PA21 PA22 PA23 PA24 PA25 PB00 PB01 HIF Boot Mode Function Settable by PFC Initial Function PA16/A16 PA17/A17 PA18/A18 PA19/A19 PA20/A20 PA21/A21 PA22/A22 PA23/A23 PA24/A24 PA25/A25 PB00/WAIT PB01/IOIS16 A00 A01 A02 A03 A04 A05 A06 A07 A08 A09 A10 A11 A12 A13 A14 A15 PA16 PA17 PA18 PA19 PA20 PA21 PA22 PA23 PA24 PA25 PB00 PB01 Function Settable by PFC PA16/A16 PA17/A17 PA18/A18 PA19/A19 PA20/A20 PA21/A21 PA22/A22 PA23/A23 PA24/A24 PA25/A25 PB00/WAIT PB01/IOIS16
Rev. 6.00 Jun. 12, 2007 Page 427 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
Not HIF Boot Mode Pin No. E12 D13 C15 E13 E15 A8 B8 A9 E14 C6 A6 C7 D6 B9 D12 D8 C9 R3 P4 M5 R4 P6 M7 N7 R7 N4 N3 N5 N6 Initial Function PB02 PB03 PB04 (WE0/DQMLL) (WE1/DQMLU/WE) PB05 PB06 RD RDWR PB07 PB08 PB09 PB10 PB11 PB12 CS0 PB13 PC00 PC01 PC02 PC03 PC04 PC05 PC06 PC07 PC08 PC09 PC10 PC11
HIF Boot Mode Function Settable by PFC Initial Function PB02/CKE PB03/CAS PB04/RAS PB05/ICIORD PB06/ICIOWR PB07/CE2B PB08/(CS6B/CE1B) PB09/CE2A PB10/(CS5B/CE1A) PB11/CS4 PB12/CS3 PB13/BS PC00/MII_RXD0 PC01/MII_RXD1 PC02/MII_RXD2 PC03/MII_RXD3 PC04/MII_TXD0 PC05/MII_TXD1 PC06/MII_TXD2 PC07/MII_TXD3 PC08/RX_DV PC09/RX_ER PC10/RX_CLK PC11/TX_ER PB02 PB03 PB04 (WE0/DQMLL) (WE1/DQMLU/WE) PB05 PB06 RD RDWR PB07 PB08 PB09 PB10 PB11 PB12 CS0 PB13 PC00 PC01 PC02 PC03 PC04 PC05 PC06 PC07 PC08 PC09 PC10 PC11 Function Settable by PFC PB02/CKE PB03/CAS PB04/RAS PB05/ICIORD PB06/ICIOWR PB07/CE2B PB08/(CS6B/CE1B) PB09/CE2A PB10/(CS5B/CE1A) PB11/CS4 PB12/CS3 PB13/BS PC00/MII_RXD0 PC01/MII_RXD1 PC02/MII_RXD2 PC03/MII_RXD3 PC04/MII_TXD0 PC05/MII_TXD1 PC06/MII_TXD2 PC07/MII_TXD3 PC08/RX_DV PC09/RX_ER PC10/RX_CLK PC11/TX_ER
Rev. 6.00 Jun. 12, 2007 Page 428 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
Not HIF Boot Mode Pin No. P7 R6 R8 P3 P2 P1 M6 M9 P8 D1 E4 D2 C1 D3 C2 C3 B2 N2 M4 N1 M3 L4 L2 L1 L3 E3 K3 K4 J2 Initial Function PC12 PC13 PC14 PC15 PC16 PC17 PC18 PC19 PC20 PD0 PD1 PD2 PD3 PD4 PD5 PD6 PD7 PE00 PE01 PE02 HIFMD PE04 PE05 PE06 PE07 PE08 PE09 PE10 PE11
HIF Boot Mode Function Settable by PFC Initial Function PC12/TX_EN PC13/TX_CLK PC14/COL PC15/CRS PC16/MDIO PC17/MDC PC18/LNKSTA PC19/EXOUT PC20/WOL PD0/IRQ0 PD1/IRQ1 PD2/IRQ2/TxD1 PD3/IRQ3/RxD1 PD4/IRQ4/SCK1 PD5/IRQ5/TxD2 PD6/IRQ6/RxD2 PD7/IRQ7/SCK2 PE00/HIFEBL PE01/HIFRDY PE02/HIFDREQ PE03/HIFMD PE04/HIFINT PE05/HIFRD PE06/HIFWR PE07/HIFRS PE08/HIFCS PE09/HIFD00 PE10/HIFD01 PE11/HIFD02 PC12 PC13 PC14 PC15 PC16 PC17 PC18 PC19 PC20 PD0 PD1 PD2 PD3 PD4 PD5 PD6 PD7 HIFEBL HIFRDY HIFDREQ HIFMD HIFINT HIFRD HIFWR HIFRS HIFCS HIFD00 HIFD01 HIFD02 Function Settable by PFC PC12/TX_EN PC13/TX_CLK PC14/COL PC15/CRS PC16/MDIO PC17/MDC PC18/LNKSTA PC19/EXOUT PC20/WOL PD0/IRQ0 PD1/IRQ1 PD2/IRQ2/TxD1 PD3/IRQ3/RxD1 PD4/IRQ4/SCK1 PD5/IRQ5/TxD2 PD6/IRQ6/RxD2 PD7/IRQ7/SCK2 PE00/HIFEBL PE01/HIFRDY PE02/HIFDREQ PE03/HIFMD PE04/HIFINT PE05/HIFRD PE06/HIFWR PE07/HIFRS PE08/HIFCS PE09/HIFD00 PE10/HIFD01 PE11/HIFD02
Rev. 6.00 Jun. 12, 2007 Page 429 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
Not HIF Boot Mode Pin No. J1 J3 J4 H2 H1 G2 G1 G3 G4 F2 F1 F3 F4 L12 L13 L14 L15 K12 K13 K15 K14 F13 F12 G14 G15 H14 H15 H13 H12 Initial Function PE12 PE13 PE14 PE15 PE16 PE17 PE18 PE19 PE20 PE21 PE22 PE23 PE24 D00 D01 D02 D03 D04 D05 D06 D07 D08 D09 D10 D11 D12 D13 D14 D15
HIF Boot Mode Function Settable by PFC Initial Function PE12/HIFD03 PE13/HIFD04 PE14/HIFD05 PE15/HIFD06/TxD0 PE16/HIFD07/RxD0 PE17/HIFD08/SCK0 PE18/HIFD09/TxD1 PE19/HIFD10/RxD1 PE20/HIFD11/SCK1 PE21/HIFD12/RTS0 PE22/HIFD13/CTS0 PE23/HIFD14/RTS1 PE24/HIFD15/CTS1 HIFD03 HIFD04 HIFD05 HIFD06 HIFD07 HIFD08 HIFD09 HIFD10 HIFD11 HIFD12 HIFD13 HIFD14 HIFD15 D00 D01 D02 D03 D04 D05 D06 D07 D08 D09 D10 D11 D12 D13 D14 D15 Function Settable by PFC PE12/HIFD03 PE13/HIFD04 PE14/HIFD05 PE15/HIFD06/TxD0 PE16/HIFD07/RxD0 PE17/HIFD08/SCK0 PE18/HIFD09/TxD1 PE19/HIFD10/RxD1 PE20/HIFD11/SCK1 PE21/HIFD12/RTS0 PE22/HIFD13/CTS0 PE23/HIFD14/RTS1 PE24/HIFD15/CTS1
Rev. 6.00 Jun. 12, 2007 Page 430 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
Not HIF Boot Mode Pin No. M11 N11 R11 P11 N10 P13 R14 J15 R9 N12 R13 P9 J14 N15 R15 R12 P12 M10 N9 Initial Function TRST input TDO output TDI input TMS input TCK input EXTAL input XTAL output CKIO output CK_PHY output ASEMD input TESTMD input MD3 input MD2 input MD1 input MD0 input RES input NMI input MD5 input TESTOUT output
HIF Boot Mode Function Settable by PFC Initial Function TRST input TDO output TDI input TMS input TCK input EXTAL input XTAL output CKIO output CK_PHY output ASEMD input TESTMD input MD3 input MD2 input MD1 input MD0 input RES input NMI input MD5 input TESTOUT output Function Settable by PFC
Rev. 6.00 Jun. 12, 2007 Page 431 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
16.1
Register Descriptions
The PFC has the following registers. For details on the addresses of these registers and the states of these registers in each processing state, see section 20, List of Registers. * * * * * * * * * * * * * * * * * * * Port A IO register H (PAIORH) Port A control register H1 (PACRH1) Port A control register H2 (PACRH2) Port B IO register L (PBIORL) Port B control register L1 (PBCRL1) Port B control register L2 (PBCRL2) Port C IO register H (PCIORH) Port C IO register L (PCIORL) Port C control register H2 (PCCRH2) Port C control register L1 (PCCRL1) Port C control register L2 (PCCRL2) Port D IO register L (PDIORL) Port D control register L2 (PDCRL2) Port E IO register H (PEIORH) Port E IO register L (PEIORL) Port E control register H1 (PECRH1) Port E control register H2 (PECRH2) Port E control register L1 (PECRL1) Port E control register L2 (PECRL2)
Rev. 6.00 Jun. 12, 2007 Page 432 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
16.1.1
Port A IO Register H (PAIORH)
PAIORH is a 16-bit readable/writable register that selects the input/output directions of the port A pins. Bits PA25IOR to PA16IOR correspond to pins PA25 to PA16 (the pin name abbreviations for multiplexed functions are omitted). PAIORH is enabled when a port A pin functions as a general input/output (PA25 to PA16), otherwise, disabled. Setting a bit in PAIORH to 1 makes the corresponding pin function as an output and clearing a bit in PAIORH to 0 makes the pin function as an input. Bits 15 to 10 in PAIORH are reserved. These bits are always read as 0. The write value should always be 0. The initial value of PAIORH is H'0000. 16.1.2 Port A Control Register H1 and H2 (PACRH1 and PACRH2)
PACRH1 and PACRH2 are 16-bit readable/writable registers that select the pin functions for the multiplexed port A pins. * PACRH1
Bit 15 to 3 Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 2 PA25MD0 0 R/W PA25 Mode Selects the function of pin PA25/A25. 0: PA25 input/output (port) 1: A25 output (BSC) 1 0 R Reserved This bit is always read as 0. The write value should always be 0. 0 PA24MD0 0 R/W PA24 Mode Selects the function of pin PA24/A24. 0: PA24 input/output (port) 1: A24 output (BSC)
Rev. 6.00 Jun. 12, 2007 Page 433 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
* PACRH2
Bit 15 Bit Name Initial Value 0 R/W R Description Reserved This bit is always read as 0. The write value should always be 0. 14 PA23MD0 0 R/W PA23 Mode Selects the function of pin PA23/A23. 0: PA23 input/output (port) 1: A23 output (BSC) 13 0 R Reserved This bit is always read as 0. The write value should always be 0. 12 PA22MD0 0 R/W PA22 Mode Selects the function of pin PA22/A22. 0: PA22 input/output (port) 1: A22 output (BSC) 11 0 R Reserved This bit is always read as 0. The write value should always be 0. 10 PA21MD0 0 R/W PA21 Mode Selects the function of pin PA21/A21. 0: PA21 input/output (port) 1: A21 output (BSC) 9 0 R Reserved This bit is always read as 0. The write value should always be 0. 8 PA20MD0 0 R/W PA20 Mode Selects the function of pin PA20/A20. 0: PA20 input/output (port) 1: A20 output (BSC) 7 0 R Reserved This bit is always read as 0. The write value should always be 0.
Rev. 6.00 Jun. 12, 2007 Page 434 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
Bit 6
Bit Name PA19MD0
Initial Value 0
R/W R/W
Description PA19 Mode Selects the function of pin PA19/A19. 0: PA19 input/output (port) 1: A19 output (BSC)
5
0
R
Reserved This bit is always read as 0. The write value should always be 0.
4
PA18MD0
0
R/W
PA18 Mode Selects the function of pin PA18/A18. 0: PA18 input/output (port) 1: A18 output (BSC)
3
0
R
Reserved This bit is always read as 0. The write value should always be 0.
2
PA17MD0
0
R/W
PA17 Mode Selects the function of pin PA17/A17. 0: PA17 input/output (port) 1: A17 output (BSC)
1
0
R
Reserved This bit is always read as 0. The write value should always be 0.
0
PA16MD0
0
R/W
PA16 Mode Selects the function of pin PA16/A16. 0: PA16 input/output (port) 1: A16 output (BSC)
Rev. 6.00 Jun. 12, 2007 Page 435 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
16.1.3
Port B IO Register L (PBIORL)
PBIORL is a 16-bit readable/writable register that selects the input/output directions of the port B pins. Bits PB13IOR to PB0IOR correspond to pins PB13 to PB00 (the pin name abbreviations for multiplexed functions are omitted). PBIORL is enabled when a port B pin functions as a general input/output (PB13 to PB00), otherwise, disabled. Setting a bit in PBIORL to 1 makes the corresponding pin function as an output and clearing a bit in PBIORL to 0 makes the pin function as an input. Bits 15 and 14 in PBIORL are reserved. These bits are always read as 0. The write value should always be 0. The initial value of PAIBRL is H'0000. 16.1.4 Port B Control Register L1 and L2 (PBCRL1 and PBCRL2)
PBCRL1 and PBCRL2 are 16-bit readable/writable registers that select the pin functions for the multiplexed port B pins. * PBCRL1
Bit 15 to 11 Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 10 PB13MD0 0 R/W PB13 Mode Selects the function of pin PB13/BS. 0: PB13 input/output (port) 1: BS output (BSC) 9 0 R Reserved This bit is always read as 0. The write value should always be 0. 8 PB12MD0 0 R/W PB12 Mode Selects the function of pin PB12/CS3. 0: PB12 input/output (port) 1: CS3 output (BSC)
Rev. 6.00 Jun. 12, 2007 Page 436 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
Bit 7
Bit Name
Initial Value 0
R/W R
Description Reserved This bit is always read as 0. The write value should always be 0.
6
PB11MD0
0
R/W
PB11 Mode Selects the function of pin PB11/CS4. 0: PB11 input/output (port) 1: CS4 output (BSC)
5
0
R
Reserved This bit is always read as 0. The write value should always be 0.
4
PB10MD0
0
R/W
PB10 Mode Selects the function of pin PB10/CS5B/CE1A. 0: PB10 input/output (port) 1: CS5B/CE1A output (BSC)
3
0
R
Reserved This bit is always read as 0. The write value should always be 0.
2
PB9MD0
0
R/W
PB9 Mode Selects the function of pin PB09/CE2A. 0: PB09 input/output (port) 1: CE2A output (BSC)
1
0
R
Reserved This bit is always read as 0. The write value should always be 0.
0
PB8MD0
0
R/W
PB8 Mode Selects the function of pin PB08/CS6B/CE1B. 0: PB13 input/output (port) 1: CS6B/CE1B output (BSC)
Rev. 6.00 Jun. 12, 2007 Page 437 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
* PBCRL2
Bit 15 Bit Name Initial Value 0 R/W R Description Reserved This bit is always read as 0. The write value should always be 0. 14 PB7MD0 0 R/W PB7 Mode Selects the function of pin PB07/CE2B. 0: PB07 input/output (port) 1: CE2B output (BSC) 13 0 R Reserved This bit is always read as 0. The write value should always be 0. 12 PB6MD0 0 R/W PB6 Mode Selects the function of pin PB06/ICIOWR. 0: PB06 input/output (port) 1: ICIOWR output (BSC) 11 0 R Reserved This bit is always read as 0. The write value should always be 0. 10 PB5MD0 0 R/W PB5 Mode Selects the function of pin PB05/ICIORD. 0: PB05 input/output (port) 1: ICIORD output (BSC) 9 0 R Reserved This bit is always read as 0. The write value should always be 0. 8 PB4MD0 0 R/W PB4 Mode Selects the function of pin PB04/RAS. 0: PB04 input/output (port) 1: RAS output (BSC) 7 0 R Reserved This bit is always read as 0. The write value should always be 0.
Rev. 6.00 Jun. 12, 2007 Page 438 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
Bit 6
Bit Name PB3MD0
Initial Value 0
R/W R/W
Description PB3 Mode Selects the function of pin PB03/CAS. 0: PB03 input/output (port) 1: CAS output (BSC)
5
0
R
Reserved This bit is always read as 0. The write value should always be 0.
4
PB2MD0
0
R/W
PB2 Mode Selects the function of pin PB02/CKE. 0: PB02 input/output (port) 1: CKE output (BSC)
3
0
R
Reserved This bit is always read as 0. The write value should always be 0.
2
PB1MD0
0
R/W
PB1 Mode Selects the function of pin PB01/IOIS16. 0: PB01 input/output (port) 1: IOIS16 input (BSC)
1
0
R
Reserved This bit is always read as 0. The write value should always be 0.
0
PB0MD0
0
R/W
PB0 Mode Selects the function of pin PB00/WAIT. 0: PB00 input/output (port) 1: WAIT input (BSC)
Rev. 6.00 Jun. 12, 2007 Page 439 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
16.1.5
Port C IO Register H and L (PCIORH and PCIORL)
PCIORH and PCIORL are 16-bit readable/writable registers that select the input/output directions of the port C pins. Bits PC20IOR to PC0IOR correspond to pins PC20 to PC00 (the pin name abbreviations for multiplexed functions are omitted). PCIORH is enabled when a port C pin functions as a general input/output (PC20 to PC16), otherwise, disabled. PCIORL is enabled when a port C pin functions as a general input/output (PC15 to PC00), otherwise, disabled. Setting a bit in PCIORH and PCIORL to 1 makes the corresponding pin function as an output and clearing a bit in PCIORH and PCIORL to 0 makes the pin function as an input. Bits 15 to 5 in PCIORH are reserved. These bits are always read as 0. The write value should always be 0. The initial values of PCIORH and PCIORL are H'0000. 16.1.6 Port C Control Register H2, L1, and L2 (PCCRH2, PCCRL1, and PCCRL2)
PCCRH2, PCCRL1, and PCCRL2 are 16-bit readable/writable registers that select the pin functions for the multiplexed port C pins. * PCCRH2
Bit 15 to 9 Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 8 PC20MD0 0 R/W PC20 Mode Selects the function of pin PC20/WOL. 0: PC20 input/output (port) 1: WOL output (EtherC) 7 0 R Reserved This bit is always read as 0. The write value should always be 0.
Rev. 6.00 Jun. 12, 2007 Page 440 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
Bit 6
Bit Name PC19MD0
Initial Value 0
R/W R/W
Description PC19 Mode Selects the function of pin PC19/EXOUT. 0: PC19 input/output (port) 1: EXOUT output (EtherC)
5
0
R
Reserved This bit is always read as 0. The write value should always be 0.
4
PC18MD0
0
R/W
PC18 Mode Selects the function of pin PC18/LNKSTA. 0: PC18 input/output (port) 1: LNKSTA input (EtherC)
3
0
R
Reserved This bit is always read as 0. The write value should always be 0.
2
PC17MD0
0
R/W
PC17 Mode Selects the function of pin PC17/MDC. 0: PC17 input/output (port) 1: MDC output (EtherC)
1
0
R
Reserved This bit is always read as 0. The write value should always be 0.
0
PC16MD0
0
R/W
PC16 Mode Selects the function of pin PC16/MDIO. 0: PC16 input/output (port) 1: MDIO input/output (EtherC)
Rev. 6.00 Jun. 12, 2007 Page 441 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
* PCCRL1
Bit 15 Bit Name Initial Value 0 R/W R Description Reserved This bit is always read as 0. The write value should always be 0. 14 PC15MD0 0 R/W PC15 Mode Selects the function of pin PC15/CRS. 0: PC15 input/output (port) 1: CRS input (EtherC) 13 0 R Reserved This bit is always read as 0. The write value should always be 0. 12 PC14MD0 0 R/W PC14 Mode Selects the function of pin PC14/COL. 0: PC14 input/output (port) 1: COL input (EtherC) 11 0 R Reserved This bit is always read as 0. The write value should always be 0. 10 PC13MD0 0 R/W PC13 Mode Selects the function of pin PC13/TX_CLK. 0: PC13 input/output (port) 1: TX_CLK input (EtherC) 9 0 R Reserved This bit is always read as 0. The write value should always be 0. 8 PC12MD0 0 R/W PC12 Mode Selects the function of pin PC12/TX_EN. 0: PC12 input/output (port) 1: TX_EN output (EtherC) 7 0 R Reserved This bit is always read as 0. The write value should always be 0.
Rev. 6.00 Jun. 12, 2007 Page 442 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
Bit 6
Bit Name PC11MD0
Initial Value 0
R/W R/W
Description PC11 Mode Selects the function of pin PC11/TX_ER. 0: PC11 input/output (port) 1: TX_ER output (EtherC)
5
0
R
Reserved This bit is always read as 0. The write value should always be 0.
4
PC10MD0
0
R/W
PC10 Mode Selects the function of pin PC10/RX_CLK. 0: PC10 input/output (port) 1: RX_CLK input (EtherC)
3
0
R
Reserved This bit is always read as 0. The write value should always be 0.
2
PC9MD0
0
R/W
PC9 Mode Selects the function of pin PC09/RX_ER. 0: PC09 input/output (port) 1: RX_ER input (EtherC)
1
0
R
Reserved This bit is always read as 0. The write value should always be 0.
0
PC8MD0
0
R/W
PC8 Mode Selects the function of pin PC08/RX_DV. 0: PC08 input/output (port) 1: RX_DV input (EtherC)
Rev. 6.00 Jun. 12, 2007 Page 443 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
* PCCRL2
Bit 15 Bit Name Initial Value 0 R/W R Description Reserved This bit is always read as 0. The write value should always be 0. 14 PC7MD0 0 R/W PC7 Mode Selects the function of pin PC07/MII_TXD3. 0: PC07 input/output (port) 1: MII_TXD3 output (EtherC) 13 0 R Reserved This bit is always read as 0. The write value should always be 0. 12 PC6MD0 0 R/W PC6 Mode Selects the function of pin PC06/MII_TXD2. 0: PC06 input/output (port) 1: MII_TXD2 output (EtherC) 11 0 R Reserved This bit is always read as 0. The write value should always be 0. 10 PC5MD0 0 R/W PC5 Mode Selects the function of pin PC05/MII_TXD1. 0: PC05 input/output (port) 1: MII_TXD1 output (EtherC) 9 0 R Reserved This bit is always read as 0. The write value should always be 0. 8 PC4MD0 0 R/W PC4 Mode Selects the function of pin PC04/MII_TXD0. 0: PC04 input/output (port) 1: MII_TXD0 output (EtherC) 7 0 R Reserved This bit is always read as 0. The write value should always be 0.
Rev. 6.00 Jun. 12, 2007 Page 444 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
Bit 6
Bit Name PC3MD0
Initial Value 0
R/W R/W
Description PC3 Mode Selects the function of pin PC03/MII_RXD3. 0: PC03 input/output (port) 1: MII_RXD3 input (EtherC)
5
0
R
Reserved This bit is always read as 0. The write value should always be 0.
4
PC2MD0
0
R/W
PC2 Mode Selects the function of pin PC02/MII_RXD2. 0: PC02 input/output (port) 1: MII_RXD2 input (EtherC)
3
0
R
Reserved This bit is always read as 0. The write value should always be 0.
2
PC1MD0
0
R/W
PC1 Mode Selects the function of pin PC01/MII_RXD1. 0: PC01 input/output (port) 1: MII_RXD1 input (EtherC)
1
0
R
Reserved This bit is always read as 0. The write value should always be 0.
0
PC0MD0
0
R/W
PC0 Mode Selects the function of pin PC00/MII_RXD0. 0: PC00 input/output (port) 1: MII_RXD0 input (EtherC)
16.1.7
Port D IO Register L (PDIORL)
PDIORL is a 16-bit readable/writable register that selects the input/output directions of the port D pins. Bits PD7IOR to PD0IOR correspond to pins PD7 to PD0 (the pin name abbreviations for multiplexed functions are omitted). PDIORL is enabled when a port C pin functions as a general input/output (PD7 to PD0), otherwise, disabled.
Rev. 6.00 Jun. 12, 2007 Page 445 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
Setting a bit in PDIORL to 1 makes the corresponding pin function as an output and clearing a bit in PDIORL to 0 makes the pin function as an input. Bits 15 to 8 in PDIORL are reserved. These bits are always read as 0. The write value should always be 0. The initial value of PDIORL is H'0000. 16.1.8 Port D Control Register L2 (PDCRL2)
PDCRL2 is a 16-bit readable/writable register that selects the pin functions for the multiplexed port B pins. * PDCRL2
Bit 15 14 Bit Name PD7MD1 PD7MD0 Initial Value 0 0 R/W R/W R/W Description PD7 Mode Selects the function of pin PD7/IRQ7/SCK2. 00: PD7 input/output (port) 01: IRQ7 input (INTC) 10: SCK2 input/output (SCIF) 11: Setting prohibited 13 12 PD6MD1 PD6MD0 0 0 R/W R/W PD6 Mode Selects the function of pin PD6/IRQ6/RxD2. 00: PD6 input/output (port) 01: IRQ6 input (INTC) 10: RxD2 input (SCIF) 11: Setting prohibited 11 10 PD5MD1 PD5MD0 0 0 R/W R/W PD5 Mode Selects the function of pin PD5/IRQ5/TxD2. 00: PD5 input/output (port) 01: IRQ5 input (INTC) 10: TxD2 output (SCIF) 11: Setting prohibited
Rev. 6.00 Jun. 12, 2007 Page 446 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
Bit 9 8
Bit Name PD4MD1 PD4MD0
Initial Value 0 0
R/W R/W R/W
Description PD4 Mode Selects the function of pin PD4/IRQ4/SCK1. 00: PD4 input/output (port) 01: IRQ4 input (INTC) 10: SCK1 input/output (SCIF) 11: Setting prohibited
7 6
PD3MD1 PD3MD0
0 0
R/W R/W
PD3 Mode Selects the function of pin PD3/IRQ3/RxD1. 00: PD3 input/output (port) 01: IRQ3 input (INTC) 10: RxD1 input (SCIF) 11: Setting prohibited
5 4
PD2MD1 PD2MD0
0 0
R/W R/W
PD2 Mode Selects the function of pin PD2/IRQ2/TxD1. 00: PD2 input/output (port) 01: IRQ2 input (INTC) 10: TxD1 output (SCIF) 11: Setting prohibited
3
0
R
Reserved This bit is always read as 0. The write value should always be 0.
2
PD1MD0
0
R/W
PD1 Mode Selects the function of pin PD1/IRQ1. 0: PD1 input/output (port) 1: IRQ1 input (INTC)
1
0
R
Reserved This bit is always read as 0. The write value should always be 0.
0
PD0MD0
0
R/W
PD0 Mode Selects the function of pin PD0/IRQ0. 0: PD0 input/output (port) 1: IRQ0 input (INTC)
Rev. 6.00 Jun. 12, 2007 Page 447 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
16.1.9
Port E IO Register H and L (PEIORH and PEIORL)
PEIORH and PEIORL are 16-bit readable/writable registers that select the input/output directions of the port E pins. Bits PE24IOR to PE0IOR correspond to pins PE24 to PE00 (the pin name abbreviations for multiplexed functions are omitted). PEIORH is enabled when a port E pin functions as a general input/output (PE24 to PE16), otherwise, disabled. PEIORL is enabled when a port E pin functions as a general input/output (PE15 to PE00), otherwise, disabled. Setting a bit in PEIORH and PEIORL to 1 makes the corresponding pin function as an output and clearing a bit in PEIORH and PEIORL to 0 makes the pin function as an input. Bits 15 to 9 in PAIORH are reserved. These bits are always read as 0. The write value should always be 0. The initial values of PEIORH and PEIORL are H'0000. 16.1.10 Port E Control Register H1, H2, L1, and L2 (PECRH1, PECRH2, PECRL1, and PECRL2) PECRH1, PECRH2, PECRL1, and PECRL2 are 16-bit readable/writable registers that select the pin functions for the multiplexed port E pins. * PECRH1
Bit 15 to 2 Bit Name Initial Value R/W All 0 R Description Reserved These bits are always read as 0. The write value should always be 0. 1 0 PE24MD1 PE24MD0 0 0 (non-HIF boot mode) 0 1 (HIF boot mode) R/W R/W PE24 Mode Selects the function of pin PE24/HIFD15/CTS1. 00: PE24 input/output (port) 01: HIFD15 input/output (HIF) 10: CTS1 input (SCIF) 11: Setting prohibited
Rev. 6.00 Jun. 12, 2007 Page 448 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
* PECRH2
Bit 15 14 Bit Name PE23MD1 PE23MD0 Initial Value 0 0 (non-HIF boot mode) 0 1 (HIF boot mode) 13 12 PE22MD1 PE22MD0 0 0 (non-HIF boot mode) 0 1 (HIF boot mode) 11 10 PE21MD1 PE21MD0 0 0 (non-HIF boot mode) 0 1 (HIF boot mode) 9 8 PE20MD1 PE20MD0 0 0 (non-HIF boot mode) 0 1 (HIF boot mode) R/W R/W PE20 Mode Selects the function of pin PE20/HIFD11/SCK1. 00: PE20 input/output (port) 01: HIFD11 input/output (HIF) 10: SCK1 input/output (SCIF) 11: Setting prohibited R/W R/W PE21 Mode Selects the function of pin PE21/HIFD12/RTS0. 00: PE21 input/output (port) 01: HIFD12 input/output (HIF) 10: RTS0 output (SCIF) 11: Setting prohibited R/W R/W PE22 Mode Selects the function of pin PE22/HIFD13/CTS0. 00: PE22 input/output (port) 01: HIFD13 input/output (HIF) 10: CTS0 input (SCIF) 11: Setting prohibited R/W R/W R/W Description PE23 Mode Selects the function of pin PE23/HIFD14/RTS1. 00: PE23 input/output (port) 01: HIFD14 input/output (HIF) 10: RTS1 input (SCIF) 11: Setting prohibited
Rev. 6.00 Jun. 12, 2007 Page 449 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
Bit 7 6
Bit Name PE19MD1 PE19MD0
Initial Value 0 0 (non-HIF boot mode) 0 1 (HIF boot mode)
R/W R/W R/W
Description PE19 Mode Selects the function of pin PE19/HIFD10/RxD1. 00: PE19 input/output (port) 01: HIFD10 input/output (HIF) 10: RxD1 output (SCIF) 11: Setting prohibited
5 4
PE18MD1 PE18MD0
0 0 (non-HIF boot mode) 0 1 (HIF boot mode)
R/W R/W
PE18 Mode Selects the function of pin PE18/HIFD09/TxD1. 00: PE18 input/output (port) 01: HIFD09 input/output (HIF) 10: TxD1 output (SCIF) 11: Setting prohibited
3 2
PE17MD1 PE17MD0
0 0 (non-HIF boot mode) 0 1 (HIF boot mode)
R/W R/W
PE17 Mode Selects the function of pin PE17/HIFD08/SCK0. 00: PE17 input/output (port) 01: HIFD08 input/output (HIF) 10: SCK0 input/output (SCIF) 11: Setting prohibited
1 0
PE16MD1 PE16MD0
0 0 (non-HIF boot mode) 0 1 (HIF boot mode)
R/W R/W
PE16 Mode Selects the function of pin PE16/HIFD07/RxD0. 00: PE16 input/output (port) 01: HIFD07 input/output (HIF) 10: RxD0 input (SCIF) 11: Setting prohibited
Rev. 6.00 Jun. 12, 2007 Page 450 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
* PECRL1
Bit 15 14 Bit Name PE15MD1 PE15MD0 Initial Value 0 0 (non-HIF boot mode) 0 1 (HIF boot mode) 13 0 R Reserved This bit is always read as 0. The write value should always be 0. 12 PE14MD0 0 (non-HIF boot mode) 1 (HIF boot mode) 11 0 R Reserved This bit is always read as 0. The write value should always be 0. 10 PE13MD0 0 (non-HIF boot mode) 1 (HIF boot mode) 9 0 R Reserved This bit is always read as 0. The write value should always be 0. R/W PE13 Mode Selects the function of pin PE13/HIFD04. 0: PE13 input/output (port) 1: HIFD04 input/output (HIF) R/W PE14 Mode Selects the function of pin PE14/HIFD05. 0: PE14 input/output (port) 1: HIFD05 input/output (HIF) R/W R/W R/W Description PE15 Mode Selects the function of pin PE15/HIFD06/TxD0. 00: PE15 input/output (port) 01: HIFD06 input/output (HIF) 10: TxD0 output (SCIF) 11: Setting prohibited
Rev. 6.00 Jun. 12, 2007 Page 451 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
Bit 8
Bit Name PE12MD0
Initial Value 0 (non-HIF boot mode) 1 (HIF boot mode)
R/W R/W
Description PE12 Mode Selects the function of pin PE12/HIFD03. 0: PE12 input/output (port) 1: HIFD03 input/output (HIF)
7
0
R
Reserved This bit is always read as 0. The write value should always be 0.
6
PE11MD0
0 (non-HIF boot mode) 1 (HIF boot mode)
R/W
PE11 Mode Selects the function of pin PE11/HIFD02. 0: PE11 input/output (port) 1: HIFD02 input/output (HIF)
5
0
R
Reserved This bit is always read as 0. The write value should always be 0.
4
PE10MD0
0 (non-HIF boot mode) 1 (HIF boot mode)
R/W
PE10 Mode Selects the function of pin PE10/HIFD01. 0: PE10 input/output (port) 1: HIFD01 input/output (HIF)
3
0
R
Reserved This bit is always read as 0. The write value should always be 0.
2
PE9MD0
0 (non-HIF boot mode) 1 (HIF boot mode)
R/W
PE9 Mode Selects the function of pin PE09/HIFD00. 0: PE09 input/output (port) 1: HIFD00 input/output (HIF)
Rev. 6.00 Jun. 12, 2007 Page 452 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
Bit 1
Bit Name
Initial Value 0
R/W R
Description Reserved This bit is always read as 0. The write value should always be 0.
0
PE8MD0
0 (non-HIF boot mode) 1 (HIF boot mode)
R/W
PE8 Mode Selects the function of pin PE08/HIFCS. 0: PE08 input/output (port) 1: HIFCS input (HIF)
* PECRL2
Bit 15 Bit Name Initial Value 0 R/W R Description Reserved This bit is always read as 0. The write value should always be 0. 14 PE7MD0 0 (non-HIF boot mode) 1 (HIF boot mode) 13 0 R Reserved This bit is always read as 0. The write value should always be 0. 12 PE6MD0 0 (non-HIF boot mode) 1 (HIF boot mode) 11 0 R Reserved This bit is always read as 0. The write value should always be 0. R/W PE6 Mode Selects the function of pin PE06/HIFWR. 0: PE06 input/output (port) 1: HIFWR input (HIF) R/W PE7 Mode Selects the function of pin PE07/HIFRS. 0: PE07 input/output (port) 1: HIFRS input (HIF)
Rev. 6.00 Jun. 12, 2007 Page 453 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
Bit 10
Bit Name PE5MD0
Initial Value 0 (non-HIF boot mode) 1 (HIF boot mode)
R/W R/W
Description PE5 Mode Selects the function of pin PE05/HIFRD. 0: PE05 input/output (port) 1: HIFRD input (HIF)
9
0
R
Reserved This bit is always read as 0. The write value should always be 0.
8
PE4MD0
0 (non-HIF boot mode) 1 (HIF boot mode)
R/W
PE4 Mode Selects the function of pin PE04/HIFINT. 0: PE04 input/output (port) 1: HIFINT output (HIF)
7
0
R
Reserved This bit is always read as 0. The write value should always be 0.
6
PE3MD0
1
R/W
PE3 Mode Selects the function of pin PE03/HIFMD. 0: PE03 input/output (port) 1: HIFMD input (HIF)
5
0
R
Reserved This bit is always read as 0. The write value should always be 0.
4
PE2MD0
0 (non-HIF boot mode) 1 (HIF boot mode)
R/W
PE2 Mode Selects the function of pin PE02/HIFDREQ. 0: PE02 input/output (port) 1: HIFDREQ output (HIF)
Rev. 6.00 Jun. 12, 2007 Page 454 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
Bit 3
Bit Name
Initial Value 0
R/W R
Description Reserved This bit is always read as 0. The write value should always be 0.
2
PE1MD0
0 (non-HIF boot mode) 1 (HIF boot mode)
R/W
PE1 Mode Selects the function of pin PE01/HIFRDY. 0: PE01 input/output (port) 1: HIFRDY output (HIF)
1
0
R
Reserved This bit is always read as 0. The write value should always be 0.
0
PE0MD0
0 (non-HIF boot mode) 1 (HIF boot mode)
R/W
PE0 Mode Selects the function of pin PE00/HIFEBL. 0: PE00 input/output (port) 1: HIFEBL input (HIF)
Rev. 6.00 Jun. 12, 2007 Page 455 of 610 REJ09B0131-0600
Section 16 Pin Function Controller (PFC)
Rev. 6.00 Jun. 12, 2007 Page 456 of 610 REJ09B0131-0600
Section 17 I/O Ports
Section 17 I/O Ports
This LSI has 26 ports (ports A, B, C, D, and E). Port A, port B, port C, port D, and port E are 10bit, 14-bit, 21-bit, 8-bit, and 25-bit I/O port, respectively. The pins of each port are multiplexed with other functions. The pin function controller (PFC) handles the selection of multiplex pin functions. Each port has a data register to store data of pin.
17.1
Port A
Port A of this LSI is an I/O port with ten pins as shown in figure 17.1.
PA16 (input/output)/A16 (output) PA17 (input/output)/A17 (output) PA18 (input/output)/A18 (output) PA19 (input/output)/A19 (output) Port A PA20 (input/output)/A20 (output) PA21 (input/output)/A21 (output) PA22 (input/output)/A22 (output) PA23 (input/output)/A23 (output) PA24 (input/output)/A24 (output) PA25 (input/output)/A25 (output)
Figure 17.1 Port A 17.1.1 Register Description
Port A is a 10-bit I/O port that has a following register. For details on the address of this register and the states of this register in each processing state, see section 20, List of Registers. * Port A data register H (PADRH) 17.1.2 Port A Data Register H (PADRH)
PADRH is a 16-bit readable/writable register which stores data for port A. Bits PA25DR to PA16DR correspond to pins PA25 to PA16. (Description of multiplexed functions is omitted.)
Rev. 6.00 Jun. 12, 2007 Page 457 of 610 REJ09B0131-0600
Section 17 I/O Ports
When the pin function is general output port, if the value is written to PADRH, the value is output from the pin; if PADRH is read, the value written to the register is directly read regardless of the pin state. When the pin function is general input port, not the value of register but pin state is directly read if PADRH is read. Data can be written to PADRH but no effect on the pin state. Table 17.1 shows the reading/writing function of the port A data register H.
Bit Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 9 8 7 6 5 4 3 2 1 0 PA25DR PA24DR PA23DR PA22DR PA21DR PA20DR PA19DR PA18DR PA17DR PA16DR 0 0 0 0 0 0 0 0 0 0 R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W See table 17.1.
15 to 10
Table 17.1 Port A Data Register H (PADRH) Read/Write Operation * Bits 9 to 0 in PADRH
Pin Function General input General output Other functions PAIORH 0 1 * Read Pin state PADRH value PADRH value Write Data can be written to PADRH but no effect on the pin state. Written value is output from the pin. Data can be written to PADRH but no effect on the pin state.
Rev. 6.00 Jun. 12, 2007 Page 458 of 610 REJ09B0131-0600
Section 17 I/O Ports
17.2
Port B
Port B of this LSI is an I/O port with 14 pins as shown in figure 17.2.
PB00 (input/output)/WAIT (input) PB01 (input/output)/IOIS16 (input) PB02 (input/output)/CKE (output) PB03 (input/output)/CAS (output) PB04 (input/output)/RAS (output) PB05 (input/output)/ICIORD (output) PB06 (input/output)/ICIOWR (output) Port B PB07 (input/output)/CE2B (output) PB08 (input/output)/CS6B (output)/CE1B (output) PB09 (input/output)/CE2A (output) PB10 (input/output)/CS5B (output)/CE1A (output) PB11 (input/output)/CS4 (output) PB12 (input/output)/CS3 (output) PB13 (input/output)/BS (output)
Figure 17.2 Port B 17.2.1 Register Description
Port B is a 14-bit I/O port that has a following register. For details on the address of this register and the states of this register in each processing state, see section 20, List of Registers. * Port B data register L (PBDRL) 17.2.2 Port B Data Register L (PBDRL)
PBDRL is a 16-bit readable/writable register which stores data for port B. Bits PB13DR to PB0DR correspond to pins PB13 to PB00. (Description of multiplexed functions is omitted.) When the pin function is general output port, if the value is written to PBDRL, the value is output from the pin; if PBDRL is read, the value written to the register is directly read regardless of the pin state.
Rev. 6.00 Jun. 12, 2007 Page 459 of 610 REJ09B0131-0600
Section 17 I/O Ports
When the pin function is general input port, not the value of register but pin state is directly read if PBDRL is read. Data can be written to PBDRL but no effect on the pin state. Table 17.2 shows the reading/writing function of the port B data register L.
Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 Bit Name PB13DR PB12DR PB11DR PB10DR PB9DR PB8DR PB7DR PB6DR PB5DR PB4DR PB3DR PB2DR PB1DR PB0DR Initial Value 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 R/W R R R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W Description Reserved These bits are always read as 0. The write value should always be 0. See table 17.2.
Table 17.2 Port B Data Register L (PBDRL) Read/Write Operation * Bits 13 to 0 in PBDRL
Pin Function General input General output Other functions PBIORL 0 1 * Read Pin state PBDRL value PBDRL value Write Data can be written to PBDRL but no effect on the pin state. Written value is output from the pin. Data can be written to PBDRL but no effect on the pin state.
Rev. 6.00 Jun. 12, 2007 Page 460 of 610 REJ09B0131-0600
Section 17 I/O Ports
17.3
Port C
Port C of this LSI is an I/O port with 21 pins as shown in figure 17.3.
PC00 (input/output)/MII_RXD0 (input) PC01 (input/output)/MII_RXD1 (input) PC02 (input/output)/MII_RXD2 (input) PC03 (input/output)/MII_RXD3 (input) PC04 (input/output)/MII_TXD0 (output) PC05 (input/output)/MII_TXD1 (output) PC06 (input/output)/MII_TXD2 (output) PC07 (input/output)/MII_TXD3 (output) PC08 (input/output)/RX_DV (input) PC09 (input/output)/RX_ER (input) Port C PC10 (input/output)/RX_CLK (input) PC11 (input/output)/TX_ER (output) PC12 (input/output)/TX_EN (output) PC13 (input/output)/TX_CLK (input) PC14 (input/output)/COL (input) PC15 (input/output)/CRS (input) PC16 (input/output)/MDIO (input/output) PC17 (input/output)/MDC (output) PC18 (input/output)/LNKSTA (input) PC19 (input/output)/EXOUT (output) PC20 (input/output)/WOL (output)
Figure 17.3 Port C
Rev. 6.00 Jun. 12, 2007 Page 461 of 610 REJ09B0131-0600
Section 17 I/O Ports
17.3.1
Register Description
Port C is a 21-bit I/O port that has the following registers. For details on the addresses of these registers and the states of these registers in each processing state, see section 20, List of Registers. * Port C data register H (PCDRH) * Port C data register L (PCDRL) 17.3.2 Port C Data Registers H and L (PCDRH and PCDRL)
PCDRH and PCDRL are 16-bit readable/writable registers that stores data for port C. Bits PC20DR to PC0DR correspond to pins PC20 to PC00. (Description of multiplexed functions is omitted.) When the pin function is general output port, if the value is written to PCDRH or PCDRL, the value is output from the pin; if PCDRH or PCDRL is read, the value written to the register is directly read regardless of the pin state. When the pin function is general input port, not the value of register but pin state is directly read if PCDRH or PCDRL is read. Data can be written to PCDRH or PCDRL but no effect on the pin state. Table 17.3 shows the reading/writing function of the port C data registers H and L. * PCDRH
Bit Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 4 3 2 1 0 PC20DR PC19DR PC18DR PC17DR PC16DR 0 0 0 0 0 R/W R/W R/W R/W R/W See table 17.3.
15 to 5
Rev. 6.00 Jun. 12, 2007 Page 462 of 610 REJ09B0131-0600
Section 17 I/O Ports
* PCDRL
Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 Bit Name PC15DR PC14DR PC13DR PC12DR PC11DR PC10DR PC9DR PC8DR PC7DR PC6DR PC5DR PC4DR PC3DR PC2DR PC1DR PC0DR Initial Value 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W Description See table 17.3.
Table 17.3 Port C Data Registers H and L (PCDRH and PCDRL) Read/Write Operation * Bits 4 to 0 in PCDRH and Bits 15 to 0 in PCDRL
Pin Function General input General output Other functions PBIORL 0 1 * Read Pin state PCDRH or PCDRL value PCDRH or PCDRL value Write Data can be written to PCDRH or PCDRL but no effect on the pin state. Written value is output from the pin. Data can be written to PCDRH or PCDRL but no effect on the pin state.
Rev. 6.00 Jun. 12, 2007 Page 463 of 610 REJ09B0131-0600
Section 17 I/O Ports
17.4
Port D
Port D of this LSI is an I/O port with eight pins as shown in figure 17.4.
PD0 (input/output)/RQ0 (input) PD1 (input/output)/IRQ1 (input) PD2 (input/output)/IRQ2 (input)/TxD1 (output) PD3 (input/output)/IRQ3 (input)/RxD1 (input) Port B PD4 (input/output)/IRQ4 (input)/SCK1 (input/output) PD5 (input/output)/IRQ5 (input)/TxD2 (output) PD6 (input/output)/IRQ6 (input)/RxD2 (input) PD7 (input/output)/IRQ7 (input)/SCK2 (input/output)
Figure 17.4 Port D 17.4.1 Register Description
Port D is an 8-bit I/O port that has a following register. For details on the address of this register and the states of this register in each processing state, see section 20, List of Registers. * Port D data register L (PDDRL) 17.4.2 Port D Data Register L (PDDRL)
PDDRL is a 16-bit readable/writable register which stores data for port D. Bits PD7DR to PD0DR correspond to pins PD7 to PD0. (Description of multiplexed functions is omitted.) When the pin function is general output port, if the value is written to PDDRL, the value is output from the pin; if PDDRL is read, the value written to the register is directly read regardless of the pin state. When the pin function is general input port, not the value of register but pin state is directly read if PDDRL is read. Data can be written to PDDRL but no effect on the pin state. Table 17.4 shows the reading/writing function of the port D data register L.
Rev. 6.00 Jun. 12, 2007 Page 464 of 610 REJ09B0131-0600
Section 17 I/O Ports
Bit
Bit Name
Initial Value All 0
R/W R
Description Reserved These bits are always read as 0. The write value should always be 0.
15 to 8
7 6 5 4 3 2 1 0
PD7DR PD6DR PD5DR PD4DR PD3DR PD2DR PD1DR PD0DR
0 0 0 0 0 0 0 0
R/W R/W R/W R/W R/W R/W R/W R/W
See table 17.4.
Table 17.4 Port D Data Register L (PDDRL) Read/Write Operation * Bits 7 to 0 in PDDRL
Pin Function General input General output Other functions PBIORL 0 1 * Read Pin state PDDRL value PDDRL value Write Data can be written to PDDRL but no effect on the pin state. Written value is output from the pin. Data can be written to PDDRL but no effect on the pin state.
Rev. 6.00 Jun. 12, 2007 Page 465 of 610 REJ09B0131-0600
Section 17 I/O Ports
17.5
Port E
Port E of this LSI is an I/O port with 25 pins as shown in figure 17.5.
PE00 (input/output)/HIFEBL (input) PE01 (input/output)/HIFRDY (output) PE02 (input/output)/HIFDREQ (output) PE03 (input/output)/HIFMD (input) PE04 (input/output)/HIFINT (output) PE05 (input/output)/HIFRD (input) PE06 (input/output)/HIFWR (input) PE07 (input/output)/HIFRS (input) PE08 (input/output)/HIFCS (input) PE09 (input/output)/HIFD00 (input/output) PE10 (input/output)/HIFD01 (input/output) PE11 (input/output)/HIFD02 (input/output) Port E PE12 (input/output)/HIFD03 (input/output) PE13 (input/output)/HIFD04 (input/output) PE14 (input/output)/HIFD05 (input/output) PE15 (input/output)/HIFD06 (input/output)/TxD0 (output) PE16 (input/output)/HIFD07 (input/output)/RxD0 (input) PE17 (input/output)/HIFD08 (input/output)/SCK0 (input/output) PE18 (input/output)/HIFD09 (input/output)/TxD1 (output) PE19 (input/output)/HIFD10 (input/output)/RxD1 (input) PE20 (input/output)/HIFD11 (input/output)/SCK1 (input/output) PE21 (input/output)/HIFD12 (input/output)/RTS0 (output) PE22 (input/output)/HIFD13 (input/output)/CTS0 (input) PE23 (input/output)/HIFD14 (input/output)/RTS1 (output) PE24 (input/output)/HIFD15 (input/output)/CTS1 (input)
Figure 17.5 Port E
Rev. 6.00 Jun. 12, 2007 Page 466 of 610 REJ09B0131-0600
Section 17 I/O Ports
17.5.1
Register Description
Port E is a 25-bit I/O port that has the following registers. For details on the addresses of these registers and the states of these registers in each processing state, see section 20, List of Registers. * Port E data register H (PEDRH) * Port E data register L (PEDRL) 17.5.2 Port E Data Registers H and L (PEDRH and PEDRL)
PEDRH and PEDRL are 16-bit readable/writable registers that store data for port E. Bits PE24DR to PE0DR correspond to pins PE24 to PE00. (Description of multiplexed functions is omitted.) When the pin function is general output port, if the value is written to PEDRH or PEDRL, the value is output from the pin; if PEDRH or PEDRL is read, the value written to the register is directly read regardless of the pin state. When the pin function is general input port, not the value of register but pin state is directly read if PEDRH or PEDRL is read. Data can be written to PEDRH or PEDRL but no effect on the pin state. Table 17.5 shows the reading/writing function of the port E data registers H and L. * PEDRH
Bit Bit Name Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 8 7 6 5 4 3 2 1 0 PE24DR PE23DR PE22DR PE21DR PE20DR PE19DR PE18DR PE17DR PE16DR 0 0 0 0 0 0 0 0 0 R/W R/W R/W R/W R/W R/W R/W R/W R/W See table 17.5.
15 to 9
Rev. 6.00 Jun. 12, 2007 Page 467 of 610 REJ09B0131-0600
Section 17 I/O Ports
* PEDRL
Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 Bit Name PE15DR PE14DR PE13DR PE12DR PE11DR PE10DR PE9DR PE8DR PE7DR PE6DR PE5DR PE4DR PE3DR PE2DR PE1DR PE0DR Initial Value 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W Description See table 17.5.
Table 17.5 Port E Data Registers H, L (PEDRH, PEDRL) Read/Write Operation * Bits 8 to 0 in PEDRH and Bits 15 to 0 in PEDRL
Pin Function General input General output Other functions PBIORL 0 1 * Read Pin state PEDRH or PEDRL value PEDRH or PEDRL value Write Data can be written to PEDRH or PEDRL but no effect on the pin state. Written value is output from the pin. Data can be written to PEDRH or PEDRL but no effect on the pin state.
Rev. 6.00 Jun. 12, 2007 Page 468 of 610 REJ09B0131-0600
Section 17 I/O Ports
17.6
Usage Note
1. When pins multiplexed with general I/O is used as output pins for other functions, these pins work as general output pins for the period of 1 x tPCYC synchronized with internal power-on reset by WDT overflow. For example, when the pin PB12/CS3 works as CS3 and the PB12DR bit in PBDRL is set to 0, the pin is driven low for the period of 1 x tPCYC and may cause memory malfunction. To prevent this, port registers that correspond to pins used for the strobe output must be set to strobe non-active level. This does not apply to the power-on reset from the RES pin. 2. Since the HIFMD pin is not initially set to function as a general port pin, it must be pulled up or down externally to fix its state. 3. When using a multiplexed pin with a function not selected with its initial value (for example, using the PB12/CS3 pin, the initial function of which is PB12, as the CS3 pin), the pin must be pulled up or down externally at least after a reset until its pin function is selected by software to fix its state.
Rev. 6.00 Jun. 12, 2007 Page 469 of 610 REJ09B0131-0600
Section 17 I/O Ports
Rev. 6.00 Jun. 12, 2007 Page 470 of 610 REJ09B0131-0600
Section 18 User Break Controller (UBC)
Section 18 User Break Controller (UBC)
The user break controller (UBC) provides functions that simplify program debugging. These functions make it easy to design an effective self-monitoring debugger, enabling the chip to debug programs without using an in-circuit emulator. Break conditions that can be set in the UBC are instruction fetch or data read/write access, data size, data contents, address value, and stop timing in the case of instruction fetch.
18.1
Features
The UBC has the following features: * The following break comparison conditions can be set. Number of break channels: two channels (channels A and B) User break can be requested as either the independent or sequential condition on channels A and B (sequential break: when channel A and channel B match with break conditions in the different bus cycles in that order, a break condition is satisfied). Address (Compares addresses 32 bits): Comparison bits are maskable in 1-bit units; user can mask addresses at lower 12 bits (4-k page), lower 10 bits (1-k page), or any size of page, etc. One of the two address buses (L-bus address (LAB) and I-bus address (IAB)) can be selected. Data (only on channel B, 32-bit maskable) One of the two data buses (logic data bus (LDB) and internal data bus (IDB)) can be selected. Bus cycle: Instruction fetch or data access Read/write Operand size: Byte, word, or longword * User break interrupt is generated upon satisfying break conditions. A user-designed user-break condition interrupt exception processing routine can be run. * In an instruction fetch cycle, it can be selected that a break is set before or after an instruction is executed. * Maximum repeat times for the break condition (only for channel B): 212 - 1 times. * Four pairs of branch source/destination buffers.
UBCS300C_0000200309000
Rev. 6.00 Jun. 12, 2007 Page 471 of 610 REJ09B0131-0600
Section 18 User Break Controller (UBC)
Figure 18.1 shows a block diagram of the UBC.
Access control
IAB
LAB Access comparator
MDB
BBRA BARA
Address comparator
BAMRA
Channel A
Access comparator
BBRB
BARB Address comparator BAMRB
Data comparator Channel B
BBRB BDMRB BETR BRSR
PC trace BRDR
Control
BRCR
LDB/IDB
CPU state signal
User break request UBC location
[Legend] BBRA: BARA: BAMRA: BBRB: BARB: BAMRB:
Break bus cycle register A Break address register A Break address mask register A Break bus cycle register B Break address register B Break address mask register B
BDRB: BDMRB: BETR: BRSR: BRDR: BRCR:
Break data register B Break data mask register B Execution times break register Branch source register Branch destination register Break control register
Figure 18.1 Block Diagram of UBC
Rev. 6.00 Jun. 12, 2007 Page 472 of 610 REJ09B0131-0600
Section 18 User Break Controller (UBC)
18.2
Register Descriptions
The user break controller has the following registers. For details on register addresses and access sizes, refer to section 20, List of Registers. * * * * * * * * * * * * Break address register A (BARA) Break address mask register A (BAMRA) Break bus cycle register A (BBRA) Break address register B (BARB) Break address mask register B (BAMRB) Break bus cycle register B (BBRB) Break data register B (BDRB) Break data mask register B (BDMRB) Break control register (BRCR) Execution times break register (BETR) Branch source register (BRSR) Branch destination register (BRDR) Break Address Register A (BARA)
18.2.1
BARA is a 32-bit readable/writable register. BARA specifies the address used for a break condition in channel A.
Bit 31 to 0 Bit Name BAA31 to BAA 0 Initial Value All 0 R/W R/W Description Break Address A Store the address on the LAB or IAB specifying break conditions of channel A.
Rev. 6.00 Jun. 12, 2007 Page 473 of 610 REJ09B0131-0600
Section 18 User Break Controller (UBC)
18.2.2
Break Address Mask Register A (BAMRA)
BAMRA is a 32-bit readable/writable register. BAMRA specifies bits masked in the break address specified by BARA.
Bit 31 to 0 Bit Name BAMA31 to BAMA 0 Initial Value All 0 R/W R/W Description Break Address Mask A Specify bits masked in the channel A break address bits specified by BARA (BAA31 to BAA0). 0: Break address bit BAAn of channel A is included in the break condition 1: Break address bit BAAn of channel A is masked and is not included in the break condition Note: n = 31 to 0
18.2.3
Break Bus Cycle Register A (BBRA)
Break bus cycle register A (BBRA) is a 16-bit readable/writable register, which specifies (1) L bus cycle or I bus cycle, (2) instruction fetch or data access, (3) read or write, and (4) operand size in the break conditions of channel A.
Bit 15 to 8 Initial Bit Name Value -- All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 7 6 CDA1 CDA0 0 0 R/W R/W L Bus Cycle/I Bus Cycle Select A Select the L bus cycle or I bus cycle as the bus cycle of the channel A break condition. 00: Condition comparison is not performed 01: The break condition is the L bus cycle 10: The break condition is the I bus cycle 11: The break condition is the L bus cycle
Rev. 6.00 Jun. 12, 2007 Page 474 of 610 REJ09B0131-0600
Section 18 User Break Controller (UBC)
Bit 5 4
Initial Bit Name Value IDA1 IDA0 0 0
R/W R/W R/W
Description Instruction Fetch/Data Access Select A Select the instruction fetch cycle or data access cycle as the bus cycle of the channel A break condition. 00: Condition comparison is not performed 01: The break condition is the instruction fetch cycle 10: The break condition is the data access cycle 11: The break condition is the instruction fetch cycle or data access cycle
3 2
RWA1 RWA0
0 0
R/W R/W
Read/Write Select A Select the read cycle or write cycle as the bus cycle of the channel A break condition. 00: Condition comparison is not performed 01: The break condition is the read cycle 10: The break condition is the write cycle 11: The break condition is the read cycle or write cycle
1 0
SZA1 SZA0
0 0
R/W R/W
Operand Size Select A Select the operand size of the bus cycle for the channel A break condition. 00: The break condition does not include operand size 01: The break condition is byte access 10: The break condition is word access 11: The break condition is longword access
18.2.4
Break Address Register B (BARB)
BARB is a 32-bit readable/writable register. BARB specifies the address used for a break condition in channel B.
Bit 31 to 0 Initial Bit Name Value BAB31 to All 0 BAB 0 R/W R/W Description Break Address B Stores an address of LAB or IAB which specifies a break condition in channel B.
Rev. 6.00 Jun. 12, 2007 Page 475 of 610 REJ09B0131-0600
Section 18 User Break Controller (UBC)
18.2.5
Break Address Mask Register B (BAMRB)
BAMRB is a 32-bit readable/writable register. BAMRB specifies bits masked in the break address specified by BARB.
Bit 31 to 0 Bit Name BAMB31 to BAMB 0 Initial Value All 0 R/W R/W Description Break Address Mask B Specifies bits masked in the break address of channel B specified by BARB (BAB31 to BAB0). 0: Break address BABn of channel B is included in the break condition 1: Break address BABn of channel B is masked and is not included in the break condition Note: n = 31 to 0
18.2.6
Break Data Register B (BDRB)
BDRB is a 32-bit readable/writable register. BDBR selects data used for a break condition in channel B.
Bit 31 to 0 Bit Name BDB31 to BDB 0 Initial Value All 0 R/W R/W Description Break Data Bit B Stores data which specifies a break condition in channel B. BDRB specifies the break data on LDB or IDB. Notes: 1. Specify an operand size when including the value of the data bus in the break condition. 2. When the byte size is selected as a break condition, the same byte data must be set in bits 15 to 8 and 7 to 0 in BDRB as the break data.
Rev. 6.00 Jun. 12, 2007 Page 476 of 610 REJ09B0131-0600
Section 18 User Break Controller (UBC)
18.2.7
Break Data Mask Register B (BDMRB)
BDMRB is a 32-bit readable/writable register. BDMRB specifies bits masked in the break data specified by BDRB.
Bit 31 to 0 Bit Name BDMB31 to BDMB 0 Initial Value All 0 R/W R/W Description Break Data Mask B Specifies bits masked in the break data of channel B specified by BDRB (BDB31 to BDB0). 0: Break data BDBn of channel B is included in the break condition 1: Break data BDBn of channel B is masked and is not included in the break condition Note: n = 31 to 0 Notes: 1. Specify an operand size when including the value of the data bus in the break condition. 2. When the byte size is selected as a break condition, the same byte data must be set in bits 15-8 and 7-0 in BDMRB as the break mask data.
18.2.8
Break Bus Cycle Register B (BBRB)
Break bus cycle register B (BBRB) is a 16-bit readable/writable register, which specifies (1) L bus cycle or I bus cycle, (2) instruction fetch or data access, (3) read or write, and (4) operand size in the break conditions of channel B.
Bit 15 to 8 Bit Name -- Initial Value All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0.
Rev. 6.00 Jun. 12, 2007 Page 477 of 610 REJ09B0131-0600
Section 18 User Break Controller (UBC)
Bit 7 6
Bit Name CDB1 CDB0
Initial Value 0 0
R/W R/W R/W
Description L Bus Cycle/I Bus Cycle Select B Select the L bus cycle or I bus cycle as the bus cycle of the channel B break condition. 00: Condition comparison is not performed 01: The break condition is the L bus cycle 10: The break condition is the I bus cycle 11: The break condition is the L bus cycle
5 4
IDB1 IDB0
0 0
R/W R/W
Instruction Fetch/Data Access Select B Select the instruction fetch cycle or data access cycle as the bus cycle of the channel B break condition. 00: Condition comparison is not performed 01: The break condition is the instruction fetch cycle 10: The break condition is the data access cycle 11: The break condition is the instruction fetch cycle or data access cycle
3 2
RWB1 RWB0
0 0
R/W R/W
Read/Write Select B Select the read cycle or write cycle as the bus cycle of the channel B break condition. 00: Condition comparison is not performed 01: The break condition is the read cycle 10: The break condition is the write cycle 11: The break condition is the read cycle or write cycle
1 0
SZB1 SZB0
0 0
R/W R/W
Operand Size Select B Select the operand size of the bus cycle for the channel B break condition. 00: The break condition does not include operand size 01: The break condition is byte access 10: The break condition is word access 11: The break condition is longword access
Rev. 6.00 Jun. 12, 2007 Page 478 of 610 REJ09B0131-0600
Section 18 User Break Controller (UBC)
18.2.9
Break Control Register (BRCR)
BRCR sets the following conditions: * Channels A and B are used in two independent channel conditions or under the sequential condition. * A break is set before or after instruction execution. * Specify whether to include the number of execution times on channel B in comparison conditions. * Specify whether to include data bus on channel B in comparison conditions. * Enable PC trace. The break control register (BRCR) is a 32-bit readable/writable register that has break conditions match flags and bits for setting a variety of break conditions.
Bit 31 to 16 Initial Bit Name Value -- All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 15
SCMFCA
0
R/W
L Bus Cycle Condition Match Flag A When the L bus cycle condition in the break conditions set for channel A is satisfied, this flag is set to 1 (not cleared to 0). In order to clear this flag, write 0 into this bit. 0: The L bus cycle condition for channel A does not match 1: The L bus cycle condition for channel A matches
14
SCMFCB
0
R/W
L Bus Cycle Condition Match Flag B When the L bus cycle condition in the break conditions set for channel B is satisfied, this flag is set to 1 (not cleared to 0). In order to clear this flag, write 0 into this bit. 0: The L bus cycle condition for channel B does not match 1: The L bus cycle condition for channel B matches
Rev. 6.00 Jun. 12, 2007 Page 479 of 610 REJ09B0131-0600
Section 18 User Break Controller (UBC)
Bit 13
Initial Bit Name Value
SCMFDA
R/W R/W
Description I Bus Cycle Condition Match Flag A When the I bus cycle condition in the break conditions set for channel A is satisfied, this flag is set to 1 (not cleared to 0). In order to clear this flag, write 0 into this bit. 0: The I bus cycle condition for channel A does not match 1: The I bus cycle condition for channel A matches
0
12
SCMFDB
0
R/W
I Bus Cycle Condition Match Flag B When the I bus cycle condition in the break conditions set for channel B is satisfied, this flag is set to 1 (not cleared to 0). In order to clear this flag, write 0 into this bit. 0: The I bus cycle condition for channel B does not match 1: The I bus cycle condition for channel B matches
11
PCTE
0
R/W
PC Trace Enable 0: Disables PC trace 1: Enables PC trace
10
PCBA
0
R/W
PC Break Select A Selects the break timing of the instruction fetch cycle for channel A as before or after instruction execution. 0: PC break of channel A is set before instruction execution 1: PC break of channel A is set after instruction execution
9, 8
--
All 0
R
Reserved These bits are always read as 0. The write value should always be 0.
7
DBEB
0
R/W
Data Break Enable B Selects whether or not the data bus condition is included in the break condition of channel B. 0: No data bus condition is included in the condition of channel B 1: The data bus condition is included in the condition of channel B
Rev. 6.00 Jun. 12, 2007 Page 480 of 610 REJ09B0131-0600
Section 18 User Break Controller (UBC)
Bit 6
Initial Bit Name Value PCBB 0
R/W R/W
Description PC Break Select B Selects the break timing of the instruction fetch cycle for channel B as before or after instruction execution. 0: PC break of channel B is set before instruction execution 1: PC break of channel B is set after instruction execution
5, 4
--
All 0
R
Reserved These bits are always read as 0. The write value should always be 0.
3
SEQ
0
R/W
Sequence Condition Select Selects two conditions of channels A and B as independent or sequential conditions. 0: Channels A and B are compared under independent conditions 1: Channels A and B are compared under sequential conditions (channel A, then channel B)
2, 1
--
All 0
R
Reserved These bits are always read as 0. The write value should always be 0.
0
ETBE
0
R/W
Number of Execution Times Break Enable Enables the execution-times break condition only on channel B. If this bit is 1 (break enable), a user break is issued when the number of break conditions matches with the number of execution times that is specified by BETR. 0: The execution-times break condition is disabled on channel B 1: The execution-times break condition is enabled on channel B
Rev. 6.00 Jun. 12, 2007 Page 481 of 610 REJ09B0131-0600
Section 18 User Break Controller (UBC)
18.2.10 Execution Times Break Register (BETR) BETR is a 16-bit readable/writable register. When the execution-times break condition of channel B is enabled, this register specifies the number of execution times to make the break. The maximum number is 212 - 1 times. Every time the break condition is satisfied, BETR is decremented by 1. A break is issued when the break condition is satisfied after BETR becomes H'0001.
Bit 15 to 12 Initial Bit Name Value -- All 0 R/W R Description Reserved These bits are always read as 0. The write value should always be 0. 11 to 0 BET11 to BET0 All 0 R/W Number of Execution Times
18.2.11 Branch Source Register (BRSR) BRSR is a 32-bit read-only register. BRSR stores bits 27 to 0 in the address of the branch source instruction. BRSR has the flag bit that is set to 1 when a branch occurs. This flag bit is cleared to 0 when BRSR is read, the setting to enable PC trace is made, or BRSR is initialized by a power-on reset. Other bits are not initialized by a power-on reset. The four BRSR registers have a queue structure and a stored register is shifted at every branch.
Bit 31 Bit Name SVF Initial Value 0 R/W R Description BRSR Valid Flag Indicates whether or not the branch source address is stored. When a branch is made, this flag is set to 1. This flag is cleared to 0 by one of the following conditions: when this flag is read from this register, when PC trace is enabled, and when a power-on reset is generated. 0: The value of BRSR register is invalid 1: The value of BRSR register is valid
Rev. 6.00 Jun. 12, 2007 Page 482 of 610 REJ09B0131-0600
Section 18 User Break Controller (UBC)
Bit 30 to 28
Bit Name --
Initial Value All 0
R/W R
Description Reserved These bits are always read as 0. The write value should always be 0.
27 to 0
BSA27 to BSA0
Undefined
R
Branch Source Address Store bits 27 to 0 of the branch source address.
18.2.12 Branch Destination Register (BRDR) BRDR is a 32-bit read-only register. BRDR stores bits 27 to 0 in the address of the branch destination instruction. BRDR has the flag bit that is set to 1 when a branch occurs. This flag bit is cleared to 0 when BRDR is read, the setting to enable PC trace is made, or BRDR is initialized by a power-on reset. Other bits are not initialized by a power-on reset. The four BRDR registers have a queue structure and a stored register is shifted at every branch.
Bit 31 Bit Name DVF Initial Value 0 R/W R Description BRDR Valid Flag Indicates whether or not the branch source address is stored. When a branch is made, this flag is set to 1. This flag is cleared to 0 by one of the following conditions: when this flag is read from this register, when PC trace is enabled, and when a power-on reset is generated. 0: The value of BRDR register is invalid 1: The value of BRDR register is valid 30 to 28 -- All 0 R Reserved These bits are always read as 0. The write value should always be 0. 27 to 0 BDA27 to BDA0
Undefined
R
Branch Destination Address Store bits 27 to 0 of the branch destination address.
Rev. 6.00 Jun. 12, 2007 Page 483 of 610 REJ09B0131-0600
Section 18 User Break Controller (UBC)
18.3
18.3.1
Operation
Flow of User Break Operation
The flow from setting of break conditions to user break exception processing is described below: 1. The break addresses are set in the break address registers (BARA and BARB). The masked addresses are set in the break address mask registers (BAMRA and BAMRB). The break data is set in the break data register (BDRB). The masked data is set in the break data mask register (BDMRB). The bus break conditions are set in the break bus cycle registers (BBRA and BBRB). There are three control bit combinations in both BBRA and BBRB: bits to select Lbus cycle or I-bus cycle, bits to select instruction fetch or data access, and bits to select read or write. No user break will be generated if one of these combinations is set to B'00. The respective conditions are set in the bits of the break control register (BRCR). Make sure to set all registers related to breaks before setting BBRA/BBRB. 2. When the break conditions are satisfied, the UBC sends a user break request to the CPU and sets the L bus condition match flag (SCMFCA or SCMFCB) and the I bus condition match flag (SCMFDA or SCMFDB) for the appropriate channel. 3. The appropriate condition match flags (SCMFCA, SCMFDA, SCMFCB, and SCMFDB) can be used to check if the set conditions match or not. The matching of the conditions sets flags. Reset the flags by writing 0 before they are used again. 4. There is a chance that the data access break and its following instruction fetch break occur around the same time, there will be only one break request to the CPU, but these two break channel match flags could be both set.
Rev. 6.00 Jun. 12, 2007 Page 484 of 610 REJ09B0131-0600
Section 18 User Break Controller (UBC)
18.3.2
Break on Instruction Fetch Cycle
1. When L bus/instruction fetch/read/word or longword is set in the break bus cycle register (BBRA/BBRB), the break condition becomes the L bus instruction fetch cycle. Whether it breaks before or after the execution of the instruction can then be selected with the PCBA/PCBB bit of the break control register (BRCR) for the appropriate channel. If an instruction fetch cycle is set as a break condition, clear LSB in the break address register (BARA/BARB) to 0. A break cannot be generated as long as this bit is set to 1. 2. If the condition is matched while a break before execution is selected, a break is generated when it is confirmed that the instruction has been fetched and it will be executed. This means this feature cannot be used on instructions fetched by overrun (instructions fetched at a branch or during an interrupt transition, but not to be executed). When this kind of break is set in the delay slot of a delayed branch instruction, the break is generated immediately before the execution of the instruction that first accepts the break. Meanwhile, a break before the execution of the instruction in a delay slot and a break after the execution of the SLEEP instruction are also prohibited. 3. When a break after execution is selected, the instruction that matches the break condition is executed and then the break is generated prior to the execution of the next instruction. As with a break before execution, this cannot be used with overrun fetch instructions. When this kind of break is set for a delayed branch instruction, a break is not generated until the first instruction at which breaks are accepted. 4. When an instruction fetch cycle is set for channel B, the break data register B (BDRB) is ignored. There is thus no need to set break data for the break of the instruction fetch cycle. 18.3.3 Break on Data Access Cycle
* The bus cycles in which L bus data access breaks occur are from instructions. * The relationship between the data access cycle address and the comparison condition for each operand size is listed in table 18.1. Table 18.1 Data Access Cycle Addresses and Operand Size Comparison Conditions
Access Size Longword Word Byte Address Compared Compares break address register bits 31 to 2 to address bus bits 31 to 2 Compares break address register bits 31 to 1 to address bus bits 31 to 1 Compares break address register bits 31 to 0 to address bus bits 31 to 0
Rev. 6.00 Jun. 12, 2007 Page 485 of 610 REJ09B0131-0600
Section 18 User Break Controller (UBC)
This means that when address H'00001003 is set in the break address register (BARA or BARB), for example, the bus cycle in which the break condition is satisfied is as follows (where other conditions are met). Longword access at H'00001000 Word access at H'00001002 Byte access at H'00001003 * When the data value is included in the break conditions on channel B: When the data value is included in the break conditions, either longword, word, or byte is specified as the operand size of the break bus cycle registers (BBRA and BBRB). In this case, a break is generated when the address conditions and data conditions both match. To specify byte data for this case, set the same data in two bytes at bits 15 to 8 and bits 7 to 0 of the break data register B (BDRB) and break data mask register B (BDMRB). When word or byte is set, bits 31 to 16 of BDRB and BDMRB are ignored. 18.3.4 Sequential Break
* By setting the SEQ bit in BRCR to 1, the sequential break is issued when a channel B break condition matches after a channel A break condition matches. A user break is not generated even if a channel B break condition matches before a channel A break condition matches. When channels A and B break conditions match at the same time, the sequential break is not issued. To clear the channel A condition match when a channel A condition match has occurred but a channel B condition match has not yet occurred in a sequential break specification, clear the SEQ bit in BRCR to 0. * In sequential break specification, the L- or I-bus can be selected and the execution times break condition can be also specified. For example, when the execution times break condition is specified, the break is generated when a channel B condition matches with BETR = H'0001 after a channel A condition has matched. 18.3.5 Value of Saved Program Counter (PC)
When a break occurs, PC is saved onto the stack. The PC value saved is as follows depending on the type of break. * When a break before execution is selected: The value of the program counter (PC) saved is the address of the instruction that matches the break condition. The fetched instruction is not executed, and a break occurs before it.
Rev. 6.00 Jun. 12, 2007 Page 486 of 610 REJ09B0131-0600
Section 18 User Break Controller (UBC)
* When a break after execution is selected: The PC value saved is the address of the instruction to be executed following the instruction in which the break condition matches. The fetched instruction is executed, and a break occurs before the execution of the next instruction. * When an address in a data access cycle is specified as a break condition: The PC value is the address of the instruction to be executed following the instruction that matched the break condition. The instruction that matched the condition is executed and the break occurs before the next instruction is executed. * When an address and data in a data access cycle are specified as a break condition: The PC value is the start address of the instruction that follows the instruction already executed when break processing started. When a data value is added to the break conditions, the break will occur before the execution of an instruction that is within two instructions of the instruction that matched the break condition. Therefore, where the break will occur cannot be specified exactly. 18.3.6 PC Trace
* Setting PCTE in BRCR to 1 enables PC traces. When branch (branch instruction, and interrupt) is generated, the branch source address and branch destination address are stored in BRSR and BRDR, respectively. * The branch source address has different values due to the kind of branch. Branch instruction The branch instruction address. Interrupt and exception The address of the instruction in which the interrupt or exception was accepted. This address is equal to the return address saved onto the stack. The start address of the interrupt or exception handling routine is stored in BRDR. The TRAPA instruction belongs to interrupt and exception above. * BRSR and BRDR have four pairs of queue structures. The top of queues is read first when the address stored in the PC trace register is read. BRSR and BRDR share the read pointer. Read BRSR and BRDR in order, the queue only shifts after BRDR is read. After switching the PCTE bit (in BRCR) off and on, the values in the queues are invalid.
Rev. 6.00 Jun. 12, 2007 Page 487 of 610 REJ09B0131-0600
Section 18 User Break Controller (UBC)
18.3.7
Usage Examples
Break Condition Specified for L Bus Instruction Fetch Cycle: * Register specifications BARA = H'00000404, BAMRA = H'00000000, BBRA = H'0054, BARB = H'00008010, BAMRB = H'00000006, BBRB = H'0054, BDRB = H'00000000, BDMRB = H'00000000, BRCR = H'00300400 Specified conditions: Channel A/channel B independent mode Channel A Address: H'00000404, Address mask: H'00000000 Bus cycle: L bus/instruction fetch (after instruction execution)/read (operand size is not included in the condition) Channel B Address: H'00008010, Address mask: H'00000006 Data: H'00000000, Data mask: H'00000000 Bus cycle: L bus/instruction fetch (before instruction execution)/read (operand size is not included in the condition) A user break occurs after an instruction of address H'00000404 is executed or before instructions of addresses H'00008010 to H'00008016 are executed. * Register specifications BARA = H'00037226, BAMRA = H'00000000, BBRA = H'0056, BARB = H'0003722E, BAMRB = H'00000000, BBRB = H'0056, BDRB = H'00000000, BDMRB = H'00000000, BRCR = H'00000008 Specified conditions: Channel A/channel B sequential mode Channel A Address: H'00037226, Address mask: H'00000000 Bus cycle: L bus/instruction fetch (before instruction execution)/read/word Channel B Address: H'0003722E, Address mask: H'00000000 Data: H'00000000, Data mask: H'00000000 Bus cycle: L bus/instruction fetch (before instruction execution)/read/word After address H'00037226 is executed, a user break occurs before an instruction of address H'0003722E is executed.
Rev. 6.00 Jun. 12, 2007 Page 488 of 610 REJ09B0131-0600
Section 18 User Break Controller (UBC)
* Register specifications BARA = H'00027128, BAMRA = H'00000000, BBRA = H'005A, BARB = H'00031415, BAMRB = H'00000000, BBRB = H'0054, BDRB = H'00000000, BDMRB = H'00000000, BRCR = H'00300000 Specified conditions: Channel A/channel B independent mode Channel A Address: H'00027128, Address mask: H'00000000 Bus cycle: L bus/instruction fetch (before instruction execution)/write/word The ASID check is not included. Channel B Address: H'00031415, Address mask: H'00000000 Data: H'00000000, Data mask: H'00000000 Bus cycle: L bus/instruction fetch (before instruction execution)/read (operand size is not included in the condition) On channel A, no user break occurs since instruction fetch is not a write cycle. On channel B, no user break occurs since instruction fetch is performed for an even address. * Register specifications BARA = H'00037226, BAMRA = H'00000000, BBRA = H'005A, BARB = H'0003722E, BAMRB = H'00000000, BBRB = H'0056, BDRB = H'00000000, BDMRB = H'00000000, BRCR = H'00000008 Specified conditions: Channel A/channel B sequential mode Channel A Address: H'00037226, Address mask: H'00000000 Bus cycle: L bus/instruction fetch (before instruction execution)/write/word Channel B Address: H'0003722E, Address mask: H'00000000 Data: H'00000000, Data mask: H'00000000 Bus cycle: L bus/instruction fetch (before instruction execution)/read/word Since instruction fetch is not a write cycle on channel A, a sequential condition does not match. Therefore, no user break occurs.
Rev. 6.00 Jun. 12, 2007 Page 489 of 610 REJ09B0131-0600
Section 18 User Break Controller (UBC)
* Register specifications BARA = H'00000500, BAMRA = H'00000000, BBRA = H'0057, BARB = H'00001000, BAMRB = H'00000000, BBRB = H'0057, BDRB = H'00000000, BDMRB = H'00000000, BRCR = H'00300001, BETR = H'0005 Specified conditions: Channel A/channel B independent mode Channel A Address: H'00000500, Address mask: H'00000000 Bus cycle: L bus/instruction fetch (before instruction execution)/read/longword The ASID check is not included. Channel B Address: H'00001000, Address mask: H'00000000 Data: H'00000000, Data mask: H'00000000 Bus cycle: L bus/instruction fetch (before instruction execution)/read/longword The number of execution-times break enable (5 times) On channel A, a user break occurs before an instruction of address H'00000500 is executed. On channel B, a user break occurs after the instruction of address H'00001000 are executed four times and before the fifth time. * Register specifications BARA = H'00008404, BAMRA = H'00000FFF, BBRA = H'0054, BARB = H'00008010, BAMRB = H'00000006, BBRB = H'0054, BDRB = H'00000000, BDMRB = H'00000000, BRCR = H'00000400 Specified conditions: Channel A/channel B independent mode Channel A Address: H'00008404, Address mask: H'00000FFF Bus cycle: L bus/instruction fetch (after instruction execution)/read (operand size is not included in the condition) Channel B Address: H'00008010, Address mask: H'00000006 Data: H'00000000, Data mask: H'00000000 Bus cycle: L bus/instruction fetch (before instruction execution)/read (operand size is not included in the condition) A user break occurs after an instruction of addresses H'00008000 to H'00008FFE is executed or before an instruction of addresses H'00008010 to H'00008016 is executed.
Rev. 6.00 Jun. 12, 2007 Page 490 of 610 REJ09B0131-0600
Section 18 User Break Controller (UBC)
Break Condition Specified for L Bus Data Access Cycle: * Register specifications BARA = H'00123456, BAMRA = H'00000000, BBRA = H'0064, BARB = H'000ABCDE, BAMRB = H'000000FF, BBRB = H'006A, BDRB = H'0000A512, BDMRB = H'00000000, BRCR = H'00000080 Specified conditions: Channel A/channel B independent mode Channel A Address: H'00123456, Address mask: H'00000000, ASID = H'80 Bus cycle: L bus/data access/read (operand size is not included in the condition) Channel B Address: H'000ABCDE, Address mask: H'000000FF Data: H'0000A512, Data mask: H'00000000 Bus cycle: L bus/data access/write/word On channel A, a user break occurs with longword read from address H'00123454, word read from address H'00123456, or byte read from address H'00123456. On channel B, a user break occurs when word H'A512 is written in addresses H'000ABC00 to H'000ABCFE. Break Condition Specified for I Bus Data Access Cycle: * Register specifications: BARA = H'00314156, BAMRA = H'00000000, BBRA = H'0094, BARB = H'00055555, BAMRB = H'00000000, BBRB = H'00A9, BDRB = H'00007878, BDMRB = H'00000F0F, BRCR = H'00000080 Specified conditions: Channel A/channel B independent mode Channel A Address: H'00314156, Address mask: H'00000000, ASID = H'80 Bus cycle: I bus/instruction fetch/read (operand size is not included in the condition) Channel B Address: H'00055555, Address mask: H'00000000, ASID = H'70 Data: H'00000078, Data mask: H'0000000F Bus cycle: I bus/data access/write/byte On channel A, a user break occurs when instruction fetch is performed for address H'00314156 in the memory space. On channel B, a user break occurs when the I bus writes byte data H'7* in address H'00055555.
Rev. 6.00 Jun. 12, 2007 Page 491 of 610 REJ09B0131-0600
Section 18 User Break Controller (UBC)
18.3.8
Usage Notes
1. The CPU can read from or write to the UBC registers via the I bus. Accordingly, during the period from executing an instruction to rewrite the UBC register till the new value is actually rewritten, the desired break may not occur. In order to know the timing when the UBC register is changed, read from the last written register. Instructions after then are valid for the newly written register value. 2. UBC cannot monitor access to the L bus and I bus in the same channel. 3. Note on specification of sequential break: A condition match occurs when a B-channel match occurs in a bus cycle after an A-channel match occurs in another bus cycle in sequential break setting. Therefore, no break occurs even if a bus cycle, in which an A-channel match and a B-channel match occur simultaneously, is set. 4. When a user break and another exception occur at the same instruction, which has higher priority is determined according to the priority levels defined in table 5.1 in section 5, Exception Handling. If an exception with higher priority occurs, the user break is not generated. Pre-execution break has the highest priority. When a post-execution break or data access break occurs simultaneously with a reexecution-type exception (including pre-execution break) that has higher priority, the reexecution-type exception is accepted, and the condition match flag is not set (see the exception in the following note). The break will occur and the condition match flag will be set only after the exception source of the re-execution-type exception has been cleared by the exception handling routine and re-execution of the same instruction has ended. When a post-execution break or data access break occurs simultaneously with a completion-type exception (TRAPA) that has higher priority, though a break does not occur, the condition match flag is set. 5. Note the following exception for the above note. If a post-execution break or data access break is satisfied by an instruction that generates a CPU address error by data access, the CPU address error is given priority over the break. Note that the UBC condition match flag is set in this case. 6. Note the following when a break occurs in a delay slot. If a pre-execution break is set at the delay slot instruction of the RTE instruction, the break does not occur until the branch destination of the RTE instruction. 7. User breaks are disabled during UBC module standby mode. Do not read from or write to the UBC registers during UBC module standby mode; the values are not guaranteed.
Rev. 6.00 Jun. 12, 2007 Page 492 of 610 REJ09B0131-0600
Section 19 User Debugging Interface (H-UDI)
Section 19 User Debugging Interface (H-UDI)
This LSI incorporates a user debugging interface (H-UDI) to provide a boundary scan function and emulator support. This section describes the boundary scan function of the H-UDI. For details on emulator functions of the H-UDI, refer to the user's manual of the relevant emulator.
19.1
Features
The H-UDI is a serial I/O interface which conforms to JTAG (Joint Test Action Group, IEEE Standard 1149.1 and IEEE Standard Test Access Port and Boundary-Scan Architecture) specifications. The H-UDI in this LSI supports a boundary scan function, and is also used for emulator connection. When using an emulator, H-UDI functions should not be used. Refer to the emulator manual for the method of connecting the emulator.
HUD0301A_000020030900
Rev. 6.00 Jun. 12, 2007 Page 493 of 610 REJ09B0131-0600
Section 19 User Debugging Interface (H-UDI)
Figure 19.1 shows a block diagram of the H-UDI.
TDI SDBPR
Shift register
SDBSR
SDIR
SDID
TDO
MUX
TCK TMS TRST TAP controller Decoder Local bus
[Legend] SDBPR: SDBSR: SDIR: SDID Bypass register Boundary scan register Instruction register :ID register
Figure 19.1 Block Diagram of H-UDI
Rev. 6.00 Jun. 12, 2007 Page 494 of 610 REJ09B0131-0600
Section 19 User Debugging Interface (H-UDI)
19.2
Input/Output Pins
Table 19.1 shows the pin configuration of the H-UDI. Table 19.1 Pin Configuration
Abbr. TCK Input/Output Input Description Serial Data Input/Output Clock Pin Data is serially supplied to the H-UDI from the data input pin (TDI) and output from the data output pin (TDO) in synchronization with this clock. TMS Input Mode Select Input Pin The state of the TAP control circuit is determined by changing this signal in synchronization with TCK. The protocol conforms to the JTAG standard (IEEE Std.1149.1). TRST Input Reset Input Pin Input is accepted asynchronously with respect to TCK, and when low, the H-UDI is reset. TRST must be low for the given period when the power is turned on regardless of using the HUDI function. This is different from the JTAG standard. For details on resets, see section 19.4.2, Reset Configuration. TDI Input Serial Data Input Pin Data transfer to the H-UDI is executed by changing this signal in synchronization with TCK. TDO Output Serial Data Output Pin Data read from the H-UDI is executed by reading this pin in synchronization with TCK. The data output timing depends on the command type set in SDIR. For details, see section 19.3.2, Instruction Register (SDIR). ASEMD Input ASE Mode Select Pin When a low level is input to the ASEMD pin, ASE mode is entered; if a high level is input, normal mode is entered. In ASE mode, the emulator functions can be used. The input level on the ASEMD pin should be held unchanged except during the RES pin assertion period.
Rev. 6.00 Jun. 12, 2007 Page 495 of 610 REJ09B0131-0600
Section 19 User Debugging Interface (H-UDI)
19.3
Register Descriptions
The H-UDI has the following registers. For details on the addresses of these registers and the states of these registers in each processing state, see section 20, List of Registers. * * * * Bypass register (SDBPR) Instruction register (SDIR) Boundary scan register (SDBSR) ID register (SDID) Bypass Register (SDBPR)
19.3.1
SDBPR is a 1-bit register that cannot be accessed by the CPU. When SDIR is set to the bypass mode, SDBPR is connected between H-UDI pins (TDI and TDO). The initial value is undefined. 19.3.2 Instruction Register (SDIR)
SDIR is a 16-bit read-only register. This register is in JTAG IDCODE in its initial state. It is initialized by TRST assertion or in the TAP test-logic-reset state, and can be written to by the HUDI irrespective of the CPU mode. Operation is not guaranteed if a reserved command is set in this register.
Bit 15 to 13 12 11 to 8 7 to 2 1 0 Bit Name TI7 to TI5 TI4 TI3 to TI0 Initial Value All 1 0 All 1 All 1 0 1 R/W R R R R R R Description Test Instruction 7 to 0 The H-UDI instruction is transferred to SDIR by a serial input from TDI. For commands, see table 19.2. Reserved These bits are always read as 1. Reserved This bit is always read as 0. Reserved This bit is always read as 1.
Rev. 6.00 Jun. 12, 2007 Page 496 of 610 REJ09B0131-0600
Section 19 User Debugging Interface (H-UDI)
Table 19.2 H-UDI Commands
Bits 15 to 8 TI7 0 0 0 0 0 0 1 1 1 TI6 0 0 0 1 1 1 0 1 1 TI5 0 1 1 0 1 1 1 1 1 TI4 0 0 1 0 0 1 0 1 TI3 TI2 TI1 TI0 Description JTAG EXTEST JTAG CLAMP JTAG HIGHZ JTAG SAMPLE/PRELOAD H-UDI reset, negate H-UDI reset, assert H-UDI interrupt JTAG IDCODE (Initial value) JTAG BYPASS Reserved
Other than above
19.3.3
Boundary Scan Register (SDBSR)
SDBSR is a 333-bit shift register, located on the PAD, for controlling the input/output pins of this LSI. The initial value is undefined. This register cannot be accessed by the CPU. Using the EXTEST, SAMPLE/PRELOAD, CLAMP, and HIGHZ commands, a boundary scan test conforming to the JTAG standard can be carried out. Table 19.3 shows the correspondence between this LSI's pins and boundary scan register bits.
Rev. 6.00 Jun. 12, 2007 Page 497 of 610 REJ09B0131-0600
Section 19 User Debugging Interface (H-UDI)
Table 19.3 External pins and Boundary Scan Register Bits
Bit Pin Name from TDI 332 331 330 329 328 327 326 325 324 323 322 321 320 319 318 317 316 315 314 313 312 311 310 309 308 307 306 305 304 PD06/IRQ6/RxD2/PD05/IRQ5/TxD2/PD04/IRQ4/SCK1/PD03/IRQ3/RxD1/PD02/IRQ2/TxD1/PD01/IRQ1/-/PD00/IRQ0/-/PE08/HIFCS PE24/HIFD15/CTS1/PE23/HIFD14/RTS1/PE22/HIFD13/CTS0/PE21/HIFD12/RTS0/PE20/HIFD11/SCK1/PE19/HIFD10/RxD1/PE18/HIFD09/TxD1/PE17/HIFD08/SCK0/PE16/HIFD07/RxD0/PE15/HIFD06/TxD0/PE14/HIFD05 PE13/HIFD04 PE12/HIFD03 PE11/HIFD02 PE10/HIFD01 PE09/HIFD00 PE07/HIFRS PE06/HIFWR PE05/HIFRD PE04/HIFINT PE03/HIFMD IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN I/O Bit 303 302 301 300 299 298 297 296 295 294 293 292 291 290 289 288 287 286 285 284 283 282 281 280 279 278 277 276 275 274 Pin Name PE02/HIFDREQ PE01/HIFRDY PE00/HIFEBL PC17/MDC/-/PC16/MDIO/-/PD06/IRQ6/RxD2/PD05/IRQ5/TxD2/PD04/IRQ4/SCK1/PD03/IRQ3/RxD1/PD02/IRQ2/TxD1/PD01/IRQ1/-/PD00/IRQ0/-/PE08/HIFCS PE24/HIFD15/CTS1/PE23/HIFD14/RTS1/PE22/HIFD13/CTS0/PE21/HIFD12/RTS0/PE20/HIFD11/SCK1/PE19/HIFD10/RxD1/PE18/HIFD09/TxD1/PE17/HIFD08/SCK0/PE16/HIFD07/RxD0/PE15/HIFD06/TxD0/PE14/HIFD05 PE13/HIFD04 PE12/HIFD03 PE11/HIFD02 PE10/HIFD01 PE09/HIFD00 PE07/HIFRS I/O IN IN IN IN IN OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT
Rev. 6.00 Jun. 12, 2007 Page 498 of 610 REJ09B0131-0600
Section 19 User Debugging Interface (H-UDI)
Bit 273 272 271 270 269 268 267 266 265 264 263 262 261 260 259 258 257 256 255 254 253 252 251 250 249 248 247 246 245 244 243 242
Pin Name PE06/HIFWR PE05/HIFRD PE04/HIFINT PE03/HIFMD PE02/HIFDREQ PE01/HIFRDY PE00/HIFEBL PC17/MDC/-/PC16/MDIO/-/PD06/IRQ6/RxD2/PD05/IRQ5/TxD2/PD04/IRQ4/SCK1/PD03/IRQ3/RxD1/PD02/IRQ2/TxD1/PD01/IRQ1/-/PD00/IRQ0/-/PE08/HIFCS PE24/HIFD15/CTS1/PE23/HIFD14/RTS1/PE22/HIFD13/CTS0/PE21/HIFD12/RTS0/PE20/HIFD11/SCK1/PE19/HIFD10/RxD1/PE18/HIFD09/TxD1/PE17/HIFD08/SCK0/PE16/HIFD07/RxD0/PE15/HIFD06/TxD0/PE14/HIFD05 PE13/HIFD04 PE12/HIFD03 PE11/HIFD02 PE10/HIFD01
I/O OUT OUT OUT OUT OUT OUT OUT OUT OUT Control Control Control Control Control Control Control Control Control Control Control Control Control Control Control Control Control Control Control Control Control Control Control
Bit 241 240 239 238 237 236 235 234 233 232 231 230 229 228 227 226 225 224 223 222 221 220 219 218 217 216 215 214 213 212 211 210
Pin Name PE09/HIFD00 PE07/HIFRS PE06/HIFWR PE05/HIFRD PE04/HIFINT PE03/HIFMD PE02/HIFDREQ PE01/HIFRDY PE00/HIFEBL PC17/MDC/-/PC16/MDIO/-/PC09/RX_ER/ PC15/CRS/ PC08/RX_DV/ PC00/MIIRXD0/ PC01/MIIRXD1/ PC02/MIIRXD2/ PC03/MIIRXD3/-/PC10/RX_CLK/-/PC18/LNKSTA PC11/TX_ER/-/PC13/TX_CLK/-/PC04/MIITXD0/-/PC05/MIITXD1/-/PC06/MIITXD2/-/PC07/MIITXD3/-/PC12/TX_EN/-/PC14/COL/-/PC20/WOL PC19/EXOUT MD3 MD5
I/O Control Control Control Control Control Control Control Control Control Control Control IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN
Rev. 6.00 Jun. 12, 2007 Page 499 of 610 REJ09B0131-0600
Section 19 User Debugging Interface (H-UDI)
Bit 209 208 207 206 205 204 203 202 201 200 199 198 197 196 195 194 193 192 191 190 189 188 187 186 185 184 183 182 181 180 179 178
Pin Name NMI TESTMD PC09/RX_ER/-/PC15/CRS/-/PC08/RX_DV/-/PC00/MIIRXD0/-/PC01/MIIRXD1/-/PC02/MIIRXD2/-/PC03/MIIRXD3/-/PC10/RX_CLK/-/PC18/LNKSTA PC11/TX_ER/-/PC13/TX_CLK/-/PC04/MIITXD0/-/PC05/MIITXD1/-/PC06/MIITXD2/-/PC07/MIITXD3/-/PC12/TX_EN/-/PC14/COL/-/PC20/WOL PC19/EXOUT TESTOUT PC09/RX_ER/-/PC15/CRS/-/PC08/RX_DV/-/PC00/MIIRXD0/-/PC01/MIIRXD1/-/PC02/MIIRXD2/-/PC03/MIIRXD3/-/PC10/RX_CLK/-/PC18/LNKSTA PC11/TX_ER/-/-
I/O IN IN OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT Control Control Control Control Control Control Control Control Control Control
Bit 177 176 175 174 173 172 171 170 169 168 167 166 165 164 163 162 161 160 159 158 157 156 155 154 153 152 151 150 149 148 147 146
Pin Name PC13/TX_CLK/-/PC04/MIITXD0/-/PC05/MIITXD1/-/PC06/MIITXD2/-/PC07/MIITXD3/-/PC12/TX_EN/-/PC14/COL/-/PC20/WOL PC19/EXOUT TESTOUT MD0 MD1 D00 D01 D02 D03 D04 D05 D06 D07 MD2 D15 D14 D13 D12 D11 D10 D09 D08 PB02/CKE PB03/CAS PB04/RAS
I/O Control Control Control Control Control Control Control Control Control Control IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN IN
Rev. 6.00 Jun. 12, 2007 Page 500 of 610 REJ09B0131-0600
Section 19 User Debugging Interface (H-UDI)
Bit 145 144 143 142 141 140 139 138 137 136 135 134 133 132 131 130 129 128 127 126 125 124 123 122 121 120 119 118 117 116 115 114
Pin Name PB12/CS3 D00 D01 D02 D03 D04 D05 D06 D07 D15 D14 D13 D12 D11 D10 D09 D08 WE0, DQMLL WE1, DQMLU, WE RDWR PB02/CKE PB03/CAS PB04/RAS PB12/CS3 A00 A01 A02 D00 D01 D02 D03 D04
I/O IN OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT Control Control Control Control Control
Bit 113 112 111 110 109 108 107 106 105 104 103 102 101 100 99 98 97 96 95 94 93 92 91 90 89 88 87 86 85 84 83 82
Pin Name D05 D06 D07 D15 D14 D13 D12 D11 D10 D09 D08 WE0, DQMLL WE1, DQMLU, WE RDWR PB02/CKE PB03/CAS PB04/RAS PB12/CS3 A00 A01 A02 PB13/BS PB11/CS4 PB00/WAIT PB05/ICIORD PB06/ICIOWR PB01/IOIS16 PB09/CE2A PB10/CS5B, CE1A PB07/CE2B PB08/CS6B, CE1B PA16/A16
I/O Control Control Control Control Control Control Control Control Control Control Control Control Control Control Control Control Control Control Control Control Control IN IN IN IN IN IN IN IN IN IN IN
Rev. 6.00 Jun. 12, 2007 Page 501 of 610 REJ09B0131-0600
Section 19 User Debugging Interface (H-UDI)
Bit 81 80 79 78 77 76 75 74 73 72 71 70 69 68 67 66 65 64 63 62 61 60 59 58 57 56 55 54 53 52 51 50
Pin Name PA17/A17 PA18/A18 PA19/A19 PA20/A20 PA21/A21 PA22/A22 PA23/A23 PA24/A24 PA25/A25 PD07/IRQ7/SCK2/A03 A04 A05 A06 A07 A08 A09 A10 A11 A12 A13 A14 A15 PB13/BS CS0 PB11/CS4 RD PB00/WAIT PB05/ICIORD PB06/ICIOWR PB01/IOIS16 PB09/CE2A
I/O IN IN IN IN IN IN IN IN IN IN OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT
Bit 49 48 47 46 45 44 43 42 41 40 39 38 37 36 35 34 33 32 31 30 29 28 27 26 25 24 23 22 21 20 19 18
Pin Name PB10/CS5B, CE1A PB07/CE2B PB08/CS6B, CE1B PA16/A16 PA17/A17 PA18/A18 PA19/A19 PA20/A20 PA21/A21 PA22/A22 PA23/A23 PA24/A24 PA25/A25 PD07/IRQ7/SCK2/A03 A04 A05 A06 A07 A08 A09 A10 A11 A12 A13 A14 A15 PB13/BS CS0 PB11/CS4 RD PB00/WAIT
I/O OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT OUT Control Control Control Control Control Control Control Control Control Control Control Control Control Control Control Control Control Control
Rev. 6.00 Jun. 12, 2007 Page 502 of 610 REJ09B0131-0600
Section 19 User Debugging Interface (H-UDI)
Bit 17 16 15 14 13 12 11 10 9 8 Note:
Pin Name PB05/ICIORD PB06/ICIOWR PB01/IOIS16 PB09/CE2A PB10/CS5B,CE1A PB07/CE2B PB08/CS6B,CE1B PA16/A16 PA17/A17 PA18/A18 *
I/O Control Control Control Control Control Control Control Control Control Control
Bit 7 6 5 4 3 2 1 0
Pin Name PA19/A19 PA20/A20 PA21/A21 PA22/A22 PA23/A23 PA24/A24 PA25/A25 PD07/IRQ7/SCK2/To TDO
I/O Control Control Control Control Control Control Control Control
Control means a low active signal. The corresponding pin is driven with an OUT value when the Control is driven low.
19.3.4
ID Register (SDID)
SDID is a 32-bit read-only register in which SDIDH and SDIDL are connected. Each register is a 16-bit that can be read by the CPU. To read this register by the H-UDI side, the contents can be read via the TDO pin when the IDCODE command is set and the TAP state is Shift-DR. Writing is disabled.
Bit 31 to 0 Bit Name DID31 to DID0 Initial Value R/W Description Device ID 31 to Device ID 0 ID register that is stipulated by JTAG. H'002B200F (initial value) for this LSI. Upper four bits may be changed according to the LSI version. SDIDH corresponds to bits 31 to 16. SDIDL corresponds to bits 15 to 0.
Refer to R description
Rev. 6.00 Jun. 12, 2007 Page 503 of 610 REJ09B0131-0600
Section 19 User Debugging Interface (H-UDI)
19.4
19.4.1
Operation
TAP Controller
Figure 19.2 shows the internal states of the TAP controller. State transitions basically conform to the JTAG standard.
1
Test-logic-reset 0 1 1 Select-DR-scan 0 0 1 Capture-DR 0 Shift-DR 1 1 Exit1-DR 0 Pause-DR 1 0 Exit2-DR 1 Update-DR 1 0 0 0 Exit2-IR 1 Update-IR 1 0 Exit1-IR 0 Pause-IR 1 0 0 1 Capture-IR 0 Shift-IR 1 1 0 Select-IR-scan 1
0
Run-test/idle
Figure 19.2 TAP Controller State Transitions Note: The transition condition is the TMS value at the rising edge of the TCK signal. The TDI value is sampled at the rising edge of the TCK signal and is shifted at the falling edge of the TCK signal. For details on change timing of the TDO value, see section 19.4.3, TDO Output Timing. The TDO pin is high impedance, except in the shift-DR and shift-IR states. A transition to the Test-Logic-Reset state is made asynchronously with TCK by driving the TRST signal 0.
Rev. 6.00 Jun. 12, 2007 Page 504 of 610 REJ09B0131-0600
Section 19 User Debugging Interface (H-UDI)
19.4.2
Reset Configuration
Table 19.4 Reset Configuration
ASEMD*1 High RES Low TRST Low High High Low High Low Low Low High High Low High LSI State Normal reset and H-UDI reset Normal reset H-UDI reset only Normal operation Reset hold*2 Normal reset H-UDI reset only Normal operation
Notes: 1. Selects to normal mode or ASE mode. ASEMD0 = high: normal mode ASEMD0 = low: ASE mode 2. In ASE mode, the reset hold state is entered by driving the RES and TRST pins low for the given time. In this state, the CPU does not start up, even if the RES pin is driven high. After that, when the TRST pin is driven high, H-UDI operation is enabled, but the CPU does not start up. The reset hold state is canceled by the following: another RES assert (power-on reset) or TRST reassert.
19.4.3
TDO Output Timing
The timing of data output from the TDO differs according to the command type set in SDIR. The timing changes at the TCK falling edge when JTAG commands (EXTEST, CLAMP, HIGHZ, SAMPLE/PRELOAD, IDCODE, and BYPASS) are set. This is a timing of the JTAG standard. When the H-UDI commands (H-UDI reset negate, H-UDI reset assert, and H-UDI interrupt) are set, the TDO signal is output at the TCK rising edge earlier than the JTAG standard by a half cycle.
Rev. 6.00 Jun. 12, 2007 Page 505 of 610 REJ09B0131-0600
Section 19 User Debugging Interface (H-UDI)
TCK
TDO (when the H-UDI command is set) TDO (when the JTAG command is set)
tTDO
tTDO
Figure 19.3 H-UDI Data Transfer Timing 19.4.4 H-UDI Reset
An H-UDI reset is generated by setting the H-UDI reset assert command in SDIR. An H-UDI reset is of the same kind as a power-on reset. An H-UDI reset is released by inputting the H-UDI reset negate command. The required time between the H-UDI reset assert command and H-UDI reset negate command is the same as time for keeping the RESETP pin low to apply a power-on reset.
SDIR
H-UDI reset assert
H-UDI reset negate
LSI internal reset
CPU state
Branch to H'A0000000
Figure 19.4 H-UDI Reset 19.4.5 H-UDI Interrupt
The H-UDI interrupt function generates an interrupt by setting an H-UDI command in SDIR. An H-UDI interrupt is an interrupt of general exceptions, resulting in a branch to an address based on the VBR value plus offset, and with return by the RTE instruction. This interrupt request has a fixed priority level of 15. H-UDI interrupts are accepted in sleep mode, but not in standby mode.
Rev. 6.00 Jun. 12, 2007 Page 506 of 610 REJ09B0131-0600
Section 19 User Debugging Interface (H-UDI)
19.5
Boundary Scan
A command can be set in SDIR by the H-UDI to place the H-UDI pins in boundary scan mode stipulated by JTAG. 19.5.1 Supported Instructions
This LSI supports the three mandatory instructions defined in the JTAG standard (BYPASS, SAMPLE/PRELOAD, and EXTEST) and three option instructions (IDCODE, CLAMP, and HIGHZ). BYPASS: The BYPASS instruction is a mandatory instruction that operates the bypass register. This instruction shortens the shift path to speed up serial data transfer involving other chips on the printed circuit board. While this instruction is executing, the test circuit has no effect on the system circuits. The upper four bits of the instruction code are 1111. SAMPLE/PRELOAD: The SAMPLE/PRELOAD instruction inputs data from this LSI's internal circuitry to the boundary scan register, outputs data from the scan path, and loads data onto the scan path. While this instruction is executed, signals input to this LSI pins are transmitted directly to the internal circuitry, and internal circuit outputs are directly output externally from the output pins. This LSI's system circuits are not affected by execution of this instruction. The upper four bits of the instruction code are 0100. In a SAMPLE operation, a snapshot of a value to be transferred from an input pin to the internal circuitry, or a value to be transferred from the internal circuitry to an output pin, is latched into the boundary scan register and read from the scan path. Snapshot latching is performed in synchronization with the rising edge of the TCK signal in the Capture-DR state. Snapshot latching does not affect normal operation of this LSI. In a PRELOAD operation, an initial value is set in the parallel output latch of the boundary scan register from the scan path prior to the EXTEST instruction. Without a PRELOAD operation, when the EXTEST instruction was executed an undefined value would be output from the output pin until completion of the initial scan sequence (transfer to the output latch) (with the EXTEST instruction, the parallel output latch value is constantly output to the output pin). EXTEST: This instruction is provided to test external circuitry when this LSI is mounted on a printed circuit board. When this instruction is executed, output pins are used to output test data (previously set by the SAMPLE/PRELOAD instruction) from the boundary scan register to the printed circuit board, and input pins are used to latch test results into the boundary scan register from the printed circuit board. If testing is carried out by using the EXTEST instruction N times, the Nth test data is scanned-in when test data (N-1) is scanned out.
Rev. 6.00 Jun. 12, 2007 Page 507 of 610 REJ09B0131-0600
Section 19 User Debugging Interface (H-UDI)
Data loaded into the output pin boundary scan register in the Capture-DR state is not used for external circuit testing (it is replaced by a shift operation). The upper four bits of the instruction code are 0000. IDCODE: A command can be set in SDIR by the H-UDI pins to place the H-UDI pins in the IDCODE mode stipulated by JTAG. When the H-UDI is initialized (TRST is asserted or TAP is in the Test-Logic-Reset state), the IDCODE mode is entered. CLAMP, HIGHZ: A command can be set in SDIR by the H-UDI pins to place the H-UDI pins in the CLAMP or HIGHZ mode stipulated by JTAG. 19.5.2 Points for Attention
* Boundary scan mode does not cover clock-related signals (EXTAL, XTAL, CKIO, and CK_PHY). * Boundary scan mode does not cover system- and E10A-related signals (RES and ASEMD). * Boundary scan mode does not cover H-UDI-related signals (TCK, TDI, TDO, TMS, and TRST). * When the EXTEST, CLAMP, and HIGHZ commands are set, fix the RES pin low. * When a boundary scan test for other than BYPASS and IDCODE is carried out, fix the ASEMD pin high.
19.6
Usage Notes
* An H-UDI command, once set, will not be modified as long as another command is not reissued from the H-UDI. If the same command is given continuously, the command must be set after a command (BYPASS, etc.) that does not affect LSI operations is once set. * Because LSI operations are suspended in standby mode, H-UDI commands are not accepted. To hold the state of the TAP before and after standby mode, the TCK signal must be high during standby mode transition. * The H-UDI is used for emulator connection. Therefore, H-UDI functions cannot be used when using an emulator.
Rev. 6.00 Jun. 12, 2007 Page 508 of 610 REJ09B0131-0600
Section 20 List of Registers
Section 20 List of Registers
The register list gives information on the on-chip I/O register addresses, how the register bits are configured, and the register states in each operating mode. The information is given as shown below. 1. Register addresses (address order) * Registers are listed from the lower allocation addresses. * Reserved addresses are indicated by in the register name column. Do not access the reserved addresses. * When registers consist of 16 or 32 bits, the addresses of the MSBs are given. * Registers are classified according to functional modules. * The numbers of Access Cycles are given. 2. Register bits * Bit configurations of the registers are listed in the same order as the register addresses. * Reserved bits are indicated by in the bit name column. * Space in the bit name field indicates that the entire register is allocated to either the counter or data. * For the registers of 16 or 32 bits, the MSB is listed first. 3. Register states in each operating mode * Register states are listed in the same order as the register addresses. * The register states shown here are for the basic operating modes. If there is a specific reset for an on-chip peripheral module, refer to the section on that on-chip peripheral module.
Rev. 6.00 Jun. 12, 2007 Page 509 of 610 REJ09B0131-0600
Section 20 List of Registers
20.1
Register Addresses (Address Order)
Entries under Access size indicates numbers of bits. The number of access cycles indicate the number of cycles of the given reference clock. B, W, and L indicate values for 8-, 16-, and 32-bit accesses, respectively. Note: Access to undefined or reserved addresses is prohibited. Since operation or continued operation is not guaranteed when these registers are accessed, do not attempt such access.
Register Name Cache Control Register 3 Port A data register H Port A IO register H Port A control register H1 Port A control register H2 Port B data register L Port B IO register L Port B control register L1 Port B control register L2 Port C data register H Port C data register L Port C IO register H Port C IO register L Port C control register H2 Port C control register L1 Port C control register L2 Port D data register L Port D IO register L Port D control register L2 Port E data register H Port E data register L Port E IO register H Port E IO register L Port E control register H1 Abbreviation No. of Bits CCR3*
2
Address H'F80000B4 H'F8050000 H'F8050004 H'F8050008 H'F805000A H'F8050012 H'F8050016 H'F805001C H'F805001E H'F8050020 H'F8050022 H'F8050024 H'F8050026 H'F805002A H'F805002C H'F805002E H'F8050032 H'F8050036 H'F805003E H'F8050040 H'F8050042 H'F8050044 H'F8050046 H'F8050048
Module Cache I/O I/O I/O I/O I/O I/O I/O I/O I/O I/O I/O I/O I/O I/O I/O I/O I/O I/O I/O I/O I/O I/O I/O
Access Size 32 8/16 8/16 8/16 8/16 8/16 8/16 8/16 8/16 8/16 8/16 8/16 8/16 8/16 8/16 8/16 8/16 8/16 8/16 8/16 8/16 8/16 8/16 8/16
32 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16
PADRH PAIORH PACRH1 PACRH2 PBDRL PBIORL PBCRL1 PBCRL2 PCDRH PCDRL PCIORH PCIORL PCCRH2 PCCRL1 PCCRL2 PDDRL PDIORL PDCRL2 PEDRH PEDRL PEIORH PEIORL PECRH1
Rev. 6.00 Jun. 12, 2007 Page 510 of 610 REJ09B0131-0600
Section 20 List of Registers
Register Name Port E control register H2 Port E control register L1 Port E control register L2 Interrupt priority register C Interrupt priority register D Interrupt priority register E Standby control register 3
Abbreviation No. of Bits PECRH2 PECRL1 PECRL2 IPRC IPRD IPRE STBCR3 16 16 16 16 16 16 8
Address H'F805004A H'F805004C H'F805004E H'F8080000 H'F8080002 H'F8080004 H'F80A0000
Module I/O I/O I/O INTC INTC INTC Powerdown mode Powerdown mode CPG H-UDI H-UDI INTC INTC INTC INTC INTC CPG Powerdown mode WDT WDT Powerdown mode SCIF_0 SCIF_0 SCIF_0 SCIF_0
Access Size 8/16 8/16 8/16 16 16 16 8
Standby control register 4
STBCR4
8
H'F80A0004
8
PHY-LSI clock frequency control register Instruction register ID register Interrupt control register 0 IRQ control register IRQ status register Interrupt priority register A Interrupt priority register B Frequency control register Standby control register
MCLKCR SDIR SDID ICR0 IRQCR IRQSR IPRA IPRB FRQCR STBCR
8 16 32 16 16 16 16 16 16 8
H'F80A000C H'F8100200 H'F8100214 H'F8140000 H'F8140002 H'F8140004 H'F8140006 H'F8140008 H'F815FF80 H'F815FF82
8/16* 16
1
16/32 8/16 8/16 8/16 8/16 8/16 16 8
Watch dog timer counter Watch dog timer control/status register Standby control register 2
WTCNT WTCSR STBCR2
8 8 8
H'F815FF84 H'F815FF86 H'F815FF88
8/16* 8/16* 8
1
1
Serial mode register_0 Bit rate register_0 Serial control register_0 Transmit FIFO data register_0
SCSMR_0 SCBRR_0 SCSCR_0 SCFTDR_0
16 8 16 8
H'F8400000 H'F8400004 H'F8400008 H'F840000C
16 8 16 8
Rev. 6.00 Jun. 12, 2007 Page 511 of 610 REJ09B0131-0600
Section 20 List of Registers
Register Name Serial status register_0 Receive FIFO data register_0 FIFO control register_0 FIFO data count register_0 Serial port register_0 Line status register_0 Serial mode register_1 Bit rate register_1 Serial control register_1 Transmit FIFO data register_1 Serial status register_1 Receive FIFO data register_1 FIFO control register_1 FIFO data count register_1 Serial Port register_1 Line status register_1 Serial mode register_2 Bit rate register_2 Serial control register_2 Transmit FIFO data register_2 Serial status register_2 Receive FIFO data register_2 FIFO control register_2 FIFO data count register_2 Serial Port register_2 Line status register_2 Compare match timer start register Compare match timer control/status register_0 Compare match counter_0 Compare match timer constant register_0
Abbreviation No. of Bits SCFSR_0 SCFRDR_0 SCFCR_0 SCFDR_0 SCSPTR_0 SCLSR_0 SCSMR_1 SCBRR_1 SCSCR_1 SCFTDR_1 SCFSR_1 SCFRDR_1 SCFCR_1 SCFDR_1 SCSPTR_1 SCLSR_1 SCSMR_2 SCBRR_2 SCSCR_2 SCFTDR_2 SCFSR_2 SCFRDR_2 SCFCR_2 SCFDR_2 SCSPTR_2 SCLSR_2 CMSTR CMCSR_0 CMCNT_0 CMCOR_0 16 8 16 16 16 16 16 8 16 8 16 8 16 16 16 16 16 8 16 8 16 8 16 16 16 16 16 16 16 16
Address H'F8400010 H'F8400014 H'F8400018 H'F840001C H'F8400020 H'F8400024 H'F8410000 H'F8410004 H'F8410008 H'F841000C H'F8410010 H'F8410014 H'F8410018 H'F841001C H'F8410020 H'F8410024 H'F8420000 H'F8420004 H'F8420008 H'F842000C H'F8420010 H'F8420014 H'F8420018 H'F842001C H'F8420020 H'F8420024 H'F84A0070 H'F84A0072 H'F84A0074 H'F84A0076
Module SCIF_0 SCIF_0 SCIF_0 SCIF_0 SCIF_0 SCIF_0 SCIF_1 SCIF_1 SCIF_1 SCIF_1 SCIF_1 SCIF_1 SCIF_1 SCIF_1 SCIF_1 SCIF_1 SCIF_2 SCIF_2 SCIF_2 SCIF_2 SCIF_2 SCIF_2 SCIF_2 SCIF_2 SCIF_2 SCIF_2 CMT CMT CMT CMT
Access Size 16 8 16 16 16 16 16 8 16 8 16 8 16 16 16 16 16 8 16 8 16 8 16 16 16 16 8/16 8/16 8/16 8/16
Rev. 6.00 Jun. 12, 2007 Page 512 of 610 REJ09B0131-0600
Section 20 List of Registers
Register Name Compare match timer control/status register_1 Compare match counter_1 Compare match timer constant register_1 HIF index register HIF general status register HIF status/control register HIF memory control register HIF internal Interrupt control register
Abbreviation No. of Bits CMCSR_1 CMCNT_1 CMCOR_1 HIFIDX HIFGSR HIFSCR HIFMCR HIFIICR 16 16 16 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32
Address H'F84A0078 H'F84A007A H'F84A007C H'F84D0000 H'F84D0004 H'F84D0008
Module CMT CMT CMT HIF HIF HIF
Access Size 8/16 8/16 8/16 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32
H'F84D000C HIF H'F84D0010 H'F84D0014 H'F84D0018 HIF HIF HIF
HIF external Interrupt control register HIFEICR HIF address register HIF data register HIFDREQ trigger register HIF bank Interrupt control register HIF boot control register Common control register Bus control register for area 0 Bus control register for area 3 Bus control register for area 4 Bus control register for area 5B Bus control register for area 6B Wait control register for area 0 Wait control register for area 3 Wait control register for area 4 Wait control register for area 5B Wait control register for area 6B SDRAM control register Refresh timer control/status register Refresh timer counter Refresh time constant register E-DMAC mode register E-DMAC transmit request register HIFADR HIFDATA HIFDTR HIFBICR HIFBCR CMNCR CS0BCR CS3BCR CS4BCR CS5BBCR CS6BBCR CS0WCR CS3WCR CS4WCR CS5BWCR CS6BWCR SDCR RTCSR RTCNT RTCOR EDMR EDTRR
H'F84D001C HIF H'F84D0020 H'F84D0024 H'F84D0040 H'F8FD0000 H'F8FD0004 HIF HIF HIF BSC BSC
H'F8FD000C BSC H'F8FD0010 H'F8FD0018 H'F8FD0020 H'F8FD0024 BSC BSC BSC BSC
H'F8FD002C BSC H'F8FD0030 H'F8FD0038 H'F8FD0040 H'F8FD0044 H'F8FD0048 BSC BSC BSC BSC BSC
H'F8FD004C BSC H'F8FD0050 H'FB000000 H'FB000004 BSC
E-DMAC 32 E-DMAC 32
Rev. 6.00 Jun. 12, 2007 Page 513 of 610 REJ09B0131-0600
Section 20 List of Registers
Register Name E-DMAC receive request register Transmit descriptor list start address register Receive descriptor list start address register EthetC/E-DMAC status register EthetC/E-DMAC status interrupt permission register Transmit/receive status copy enable register Receive missed-frame counter register Transmit FIFO threshold register FIFO depth register Receiving method control register E-DMAC operation control register Flow control FIFO threshold register Transmit Interrupt setting register Receive buffer write address register Receive descriptor fetch address register
Abbreviation No. of Bits EDRRR TDLAR RDLAR EESR EESIPR TRSCER RMFCR TFTR FDR RMCR EDOCR FCFTR TRIMD RBWAR RDFAR 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32
Address H'FB000008 H'FB00000C H'FB000010 H'FB000014 H'FB000018 H'FB00001C H'FB000020 H'FB000024 H'FB000028 H'FB00002C H'FB000030 H'FB000034 H'FB00003C H'FB000040 H'FB000044 H'FB00004C H'FB000050 H'FB000160 H'FB000164 H'FB000168 H'FB00016C H'FB000170 H'FB000174 H'FB000178 H'FB00017C H'FB000180
Module E-DMAC E-DMAC E-DMAC E-DMAC E-DMAC E-DMAC E-DMAC E-DMAC E-DMAC E-DMAC E-DMAC E-DMAC E-DMAC E-DMAC E-DMAC E-DMAC E-DMAC EtherC EtherC EtherC EtherC EtherC EtherC EtherC EtherC EtherC
Access Size 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32
Transmit buffer read address register TBRAR Transmit descriptor fetch address register EtherC mode register EtherC status register EtherC interrupt permission register PHY interface register MAC address high register MAC address low register Receive frame length register PHY status register Transmit retry over counter register TDFAR ECMR ECSR ECSIPR PIR MAHR MALR RFLR PSR TROCR
Rev. 6.00 Jun. 12, 2007 Page 514 of 610 REJ09B0131-0600
Section 20 List of Registers
Register Name Delayed collision detect counter register Lost carrier counter register Carrier not detect counter register CRC error frame receive counter register Frame receive error counter register Too-short frame receive counter register Too-long frame receive counter register Residual-bit frame counter register Multicast address frame receive counter register IPG setting register Automatic PAUSE frame set register Manual PAUSE frame set register PAUSE frame retransfer count set register Break data register B Break data mask register B Break control register Execution times break register Break address register B Break address mask register B Break bus cycle register B Branch source register Break address register A Break address mask register A Break bus cycle register A Branch destination register Cache control register 1
Abbreviation No. of Bits CDCR LCCR CNDCR CEFCR FRECR TSFRCR TLFRCR RFCR MAFCR IPGR APR MPR TPAUSER BDRB BDMRB BRCR BETR BARB BAMRB BBRB BRSR BARA BAMRA BBRA BRDR CCR1 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 16 32 32 16 32 32 32 16 32 32
Address H'FB000184 H'FB000188 H'FB00018C H'FB000194 H'FB000198 H'FB00019C H'FB0001A0 H'FB0001A4 H'FB0001A8 H'FB0001B4 H'FB0001B8
Module EtherC EtherC EtherC EtherC EtherC EtherC EtherC EtherC EtherC EtherC EtherC
Access Size 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 16 32 32 16 32 32 32 16 32 32
H'FB0001BC EtherC H'FB0001C4 H'FFFFFF90 H'FFFFFF94 H'FFFFFF98 EtherC UBC UBC UBC
H'FFFFFF9C UBC H'FFFFFFA0 UBC H'FFFFFFA4 UBC H'FFFFFFA8 UBC H'FFFFFFAC UBC H'FFFFFFB0 UBC H'FFFFFFB4 UBC H'FFFFFFB8 UBC H'FFFFFFBC UBC H'FFFFFFEC Cache
Notes: 1. The numbers of access cycles are eight bits when reading and 16 bits when writing. 2. Supported only by the SH7618A.
Rev. 6.00 Jun. 12, 2007 Page 515 of 610 REJ09B0131-0600
Section 20 List of Registers
20.2
Register Bits
Register addresses and bit names of the on-chip peripheral modules are described below. Each line covers eight bits, and 16-bit and 32-bit registers are shown as 2 or 4 lines, respectively.
Register Abbreviation CCR3* Bit Bit Bit Bit Bit Bit Bit Bit 24/16/8/0 CSIZE2 PA24DR PA16DR PA24IOR PA16IOR PA24MD0 PA20MD0 PA16MD0 PB8DR PB0DR PB8IOR PB0IOR PB12MD0 PB8MD0 PB4MD0 PB0MD0 PC16DR PC8DR PC0DR PC16IOR I/O Module Cache
31/23/15/7 30/22/14/6 29/21/13/5 28/20/12/4 27/19/11/3 26/18/10/2 25/17/9/1 CSIZE1 CSIZE0 PA22DR PA22IOR PA21DR PA21IOR PA20DR PA20IOR PA19DR PA19IOR PA18DR PA18IOR PA25DR PA17DR PA25IOR PA17IOR
PADRH
PA23DR
PAIORH
PA23IOR
PACRH1

PA25MD0 PA21MD0 PA17MD0 PB10DR PB2DR PB10IOR PB2IOR PB9DR PB1DR PB9IOR PB1IOR
PACRH2

PA23MD0 PA19MD0 PB6DR PB6IOR PB13DR PB5DR PB13IOR PB5IOR
PA22MD0 PA18MD0 PB12DR PB4DR PB12IOR PB4IOR PB11DR PB3DR PB11IOR PB3IOR
PBDRL
PB7DR
PBIORL
PB7IOR
PBCRL1

PB13MD0 PB9MD0 PB5MD0 PB1MD0 PC18DR PC10DR PC2DR PC18IOR PC17DR PC9DR PC1DR PC17IOR
PB11MD0 PB7MD0 PB3MD0 PC14DR PC6DR PC13DR PC5DR
PB10MD0 PB6MD0 PB2MD0 PC20DR PC12DR PC4DR PC20IOR PC19DR PC11DR PC3DR PC19IOR
PBCRL2

PCDRH

PCDRL
PC15DR PC7DR
PCIORH

Rev. 6.00 Jun. 12, 2007 Page 516 of 610 REJ09B0131-0600
Section 20 List of Registers
Register Abbreviation PCIORL
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit 24/16/8/0 PC8IOR PC0IOR PC20MD0 PC16MD0 PC12MD0 PC8MD0 PC4MD0 PC0MD0 PD0DR PD0IOR PD4MD0 PD0MD0 PE24DR PE16DR PE8DR PE0DR PE24IOR PE16IOR PE8IOR PE0IOR Module I/O
31/23/15/7 30/22/14/6 29/21/13/5 28/20/12/4 27/19/11/3 26/18/10/2 25/17/9/1 PC15IOR PC7IOR PC14IOR PC6IOR PC13IOR PC5IOR PC12IOR PC4IOR PC11IOR PC3IOR PC10IOR PC2IOR PC9IOR PC1IOR
PCCRH2

PC19MD0 PC15MD0 PC11MD0 PC7MD0 PC3MD0 PD6DR PD6IOR PD7MD0 PD3MD0 PE22DR PE14DR PE6DR PE22IOR PE14IOR PE6IOR PD5DR PD5IOR PD6MD1 PD2MD1 PE21DR PE13DR PE5DR PE21IOR PE13IOR PE5IOR
PC18MD0 PC14MD0 PC10MD0 PC6MD0 PC2MD0 PD4DR PD4IOR PD6MD0 PD2MD0 PE20DR PE12DR PE4DR PE20IOR PE12IOR PE4IOR PD3DR PD3IOR PD5MD1 PE19DR PE11DR PE3DR PE19IOR PE11IOR PE3IOR
PC17MD0 PC13MD0 PC9MD0 PC5MD0 PC1MD0 PD2DR PD2IOR PD5MD0 PD1MD0 PE18DR PE10DR PE2DR PE18IOR PE10IOR PE2IOR PD1DR PD1IOR PD4MD1 PE17DR PE9DR PE1DR PE17IOR PE9IOR PE1IOR
PCCRL1

PCCRL2

PDDRL
PD7DR
PDIORL
PD7IOR
PDCRL2
PD7MD1 PD3MD1
PEDRH
PE23DR
PEDRL
PE15DR PE7DR
PEIORH
PE23IOR
PEIORL
PE15IOR PE7IOR
PECRH1

PE24MD1 PE24MD0
PECRH2
PE23MD1 PE23MD0 PE22MD1 PE22MD0 PE21MD1 PE21MD0 PE20MD1 PE20MD0 PE19MD1 PE19MD0 PE18MD1 PE18MD0 PE17MD1 PE17MD0 PE16MD1 PE16MD0
PECRL1
PE15MD1 PE15MD0 PE11MD0 PE7MD0 PE3MD0
PE14MD0 PE10MD0 PE6MD0 PE2MD0
PE13MD0 PE9MD0 PE5MD0 PE1MD0
PE12MD0 PE8MD0 PE4MD0 PE0MD0
PECRL2

Rev. 6.00 Jun. 12, 2007 Page 517 of 610 REJ09B0131-0600
Section 20 List of Registers
Register Abbreviation IPRC
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit 24/16/8/0 IPRC8 IPRC0 IPRD8 IPRE8 MSTP11 MSTP19 FLDIVS0 TI0 DID24 DID16 DID8 DID0 NMIE IRQ40S IRQ00S IRQ0L IRQ0F IPRA8 IPRA0 IPRB8 IPRB0 STC0 PFC0 Powerdown mode CPG INTC Powerdown mode CPG H-UDI Module INTC
31/23/15/7 30/22/14/6 29/21/13/5 28/20/12/4 27/19/11/3 26/18/10/2 25/17/9/1 IPRC15 IPRC7 IPRC14 IPRC6 IPRD14 IPRD6 IPRE14 FLSCS0 TI6 DID30 DID22 DID14 DID6 IRQ70S IRQ30S IRQ6L IRQ6F IPRA14 IPRA6 IPRB14 IPRB6 IPRC13 IPRC5 IPRD13 IPRD5 IPRE13 TI5 DID29 DID21 DID13 DID5 IRQ61S IRQ21S IRQ5L IRQ5F IPRA13 IPRA5 IPRB13 IPRB5 IPRC12 IPRC4 IPRD12 IPRD4 IPRE12 MSTP15 MSTP23 TI4 DID28 DID20 DID12 DID4 IRQ60S IRQ20S IRQ4L IRQ4F IPRA12 IPRA4 IPRB12 IPRB4 CKOEN IPRC11 IPRC3 IPRD11 IPRE11 TI3 DID27 DID19 DID11 DID3 IRQ51S IRQ11S IRQ3L IRQ3F IPRA11 IPRA3 IPRB11 IPRB3 MDCHG IPRC10 IPRC2 IPRD10 IPRE10 MSTP13 FLDIVS2 TI2 DID26 DID18 DID10 DID2 IRQ50S IRQ10S IRQ2L IRQ2F IPRA10 IPRA2 IPRB10 IPRB2 STC2 PFC2 IPRC9 IPRC1 IPRD9 IPRE9 MSTP12 FLDIVS1 TI1 DID25 DID17 DID9 DID1 IRQ41S IRQ01S IRQ1L IRQ1F IPRA9 IPRA1 IPRB9 IPRB1 STC1 PFC1
IPRD
IPRD15 IPRD7
IPRE
IPRE15
STBCR3 STBCR4 MCLKCR SDIR
FLSCS1 TI7
SDID
DID31 DID23 DID15 DID7
ICR0
NMIL
IRQCR
IRQ71S IRQ31S
IRQSR
IRQ7L IRQ7F
IPRA
IPRA15 IPRA7
IPRB
IPRB15 IPRB7
FRQCR

STBCR
STBY
Rev. 6.00 Jun. 12, 2007 Page 518 of 610 REJ09B0131-0600
Section 20 List of Registers
Register Abbreviation WTCNT WTCSR STBCR2
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit 24/16/8/0 Bit 0 CKS0 Powerdown mode Module WDT
31/23/15/7 30/22/14/6 29/21/13/5 28/20/12/4 27/19/11/3 26/18/10/2 25/17/9/1 Bit 7 TME MSTP10 Bit 6 WT/IT MSTP9 Bit 5 Bit 4 WOVF Bit 3 IOVF Bit 2 CKS2 MSTP5 Bit 1 CKS1 MSTP4
SCSMR_0
C/A
CHR Bit 6 RIE Bit 6 PER2 TEND Bit 6 RTRG0 RTSDT CHR Bit 6 RIE Bit 6 PER2 TEND Bit 6
PE Bit 5 TE Bit 5 PER1 TDFE Bit 5 TTRG1 CTSIO PE Bit 5 TE Bit 5 PER1 TDFE Bit 5
O/E Bit 4 RE Bit 4 PER0 BRK Bit 4 TTRG0 T4 R4 CTSDT O/E Bit 4 RE Bit 4 PER0 BRK Bit 4
STOP Bit 3 REIE Bit 3 FER3 FER Bit 3 MCE T3 R3 SCKIO STOP Bit 3 REIE Bit 3 FER3 FER Bit 3
Bit 2 Bit 2 FER2 PER Bit 2 RSTRG2 TFRST T2 R2 SCKDT Bit 2 Bit 2 FER2 PER Bit 2
CKS1 Bit 1 CKE1 Bit 1 FER1 RDF Bit 1 RSTRG1 RFRST T1 R1 SPBIO CKS1 Bit 1 CKE1 Bit 1 FER1 RDF Bit 1
CKS0 Bit 0 CKE0 Bit 0 FER0 DR Bit 0 RSTRG0 LOOP T0 R0 SPBDT ORER CKS0 Bit 0 CKE0 Bit 0 FER0 DR Bit 0
SCIF_0
SCBRR_0 SCSCR_0
Bit 7 TIE
SCFTDR_0 SCFSR_0
Bit 7 PER3 ER
SCFRDR_0 SCFCR_0
Bit 7 RTRG1
SCFDR_0

SCSPTR_0
RTSIO
SCLSR_0

SCSMR_1
C/A
SCIF_1
SCBRR_1 SCSCR_1
Bit 7 TIE
SCFTDR_1 SCFSR_1
Bit 7 PER3 ER
SCFRDR_1
Bit 7
Rev. 6.00 Jun. 12, 2007 Page 519 of 610 REJ09B0131-0600
Section 20 List of Registers
Register Abbreviation SCFCR_1
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit 24/16/8/0 RSTRG0 LOOP T0 R0 SPBDT ORER CKS0 Bit 0 CKE0 Bit 0 FER0 DR Bit 0 RSTRG0 LOOP T0 R0 SPBDT SCIF_2 Module SCIF_1
31/23/15/7 30/22/14/6 29/21/13/5 28/20/12/4 27/19/11/3 26/18/10/2 25/17/9/1 RTRG1 RTRG0 RTSDT CHR Bit 6 RIE Bit 6 PER2 TEND Bit 6 RTRG0
(Reserved)
TTRG1 CTSIO PE Bit 5 TE Bit 5 PER1 TDFE Bit 5 TTRG1
(Reserved)
TTRG0 T4 R4 CTSDT O/E Bit 4 RE Bit 4 PER0 BRK Bit 4 TTRG0 T4 R4
(Reserved)
MCE T3 R3 SCKIO STOP Bit 3 REIE Bit 3 FER3 FER Bit 3 MCE T3 R3 SCKIO
RSTRG2 TFRST T2 R2 SCKDT Bit 2 Bit 2 FER2 PER Bit 2 RSTRG2 TFRST T2 R2 SCKDT
RSTRG1 RFRST T1 R1 SPBIO CKS1 Bit 1 CKE1 Bit 1 FER1 RDF Bit 1 RSTRG1 RFRST T1 R1 SPBIO
SCFDR_1

SCSPTR_1
RTSIO
SCLSR_1

SCSMR_2
C/A
SCBRR_2 SCSCR_2
Bit 7 TIE
SCFTDR_2 SCFSR_2
Bit 7 PER3 ER
SCFRDR_2 SCFCR_2
Bit 7 RTRG1
SCFDR_2

SCSPTR_2

(Reserved)
SCLSR_2

CMIE

STR1 CKS1
ORER STR0 CKS0 CMT
CMSTR

CMCSR_0
CMF
Rev. 6.00 Jun. 12, 2007 Page 520 of 610 REJ09B0131-0600
Section 20 List of Registers
Register Abbreviation CMCNT_0
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit 24/16/8/0 Bit 8 Bit 0 Bit 8 Bit 0 CKS0 Bit 8 Bit 0 Bit 8 Bit 0 BYTE0 STATUS8 STATUS0 BSEL BO AI/AD IIR HIF Module CMT
31/23/15/7 30/22/14/6 29/21/13/5 28/20/12/4 27/19/11/3 26/18/10/2 25/17/9/1 Bit 15 Bit 7 Bit 14 Bit 6 Bit 14 Bit 6 CMIE Bit 14 Bit 6 Bit 14 Bit 6 REG4 Bit 13 Bit 5 Bit 13 Bit 5 Bit 13 Bit 5 Bit 13 Bit 5 REG3 Bit 12 Bit 4 Bit 12 Bit 4 Bit 12 Bit 4 Bit 12 Bit 4 REG2 Bit 11 Bit 3 Bit 11 Bit 3 Bit 11 Bit 3 Bit 11 Bit 3 REG1 Bit 10 Bit 2 Bit 10 Bit 2 Bit 10 Bit 2 Bit 10 Bit 2 REG0 Bit 9 Bit 1 Bit 9 Bit 1 CKS1 Bit 9 Bit 1 Bit 9 Bit 1 BYTE1
CMCOR_0
Bit 15 Bit 7
CMCSR_1
CMF
CMCNT_1
Bit 15 Bit 7
CMCOR_1
Bit 15 Bit 7
HIFIDX
REG5
HIFGSR

STATUS15 STATUS14 STATUS13 STATUS12 STATUS11 STATUS10 STATUS9 STATUS7 HIFSCR HIFMCR LOCK HIFIICR IIC6 STATUS6 IIC5 STATUS5 MD1 WT IIC4 STATUS4 IIC3 STATUS3 DMD RD IIC2 STATUS2 DPOL IIC1 STATUS1 BMD EDN IIC0
Rev. 6.00 Jun. 12, 2007 Page 521 of 610 REJ09B0131-0600
Section 20 List of Registers
Register Abbreviation HIFEICR
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit 24/16/8/0 EIR A8 D24 D16 D8 D0 DTRG BIF AC HIZCNT BSC Module HIF
31/23/15/7 30/22/14/6 29/21/13/5 28/20/12/4 27/19/11/3 26/18/10/2 25/17/9/1 EIC6 EIC5 A6 D30 D22 D14 D6 EIC4 A5 D29 D21 D13 D5 EIC3 A4 D28 D20 D12 D4 MAP EIC2 A3 D27 D19 D11 D3 ENDIAN EIC1 A2 D26 D18 D10 D2 EIC0 A9 D25 D17 D9 D1 BIE HIZMEM
HIFADR
A7
HIFDATA
D31 D23 D15 D7
HIFDTR

HIFBICR

HIFBCR

CMNCR

Rev. 6.00 Jun. 12, 2007 Page 522 of 610 REJ09B0131-0600
Section 20 List of Registers
Register Abbreviation CS0BCR
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit 24/16/8/0 IWRRS0 IWRRS0 IWRRS0 IWRRS0 IWRRS0 WR1 HW0 WR1 Module BSC
31/23/15/7 30/22/14/6 29/21/13/5 28/20/12/4 27/19/11/3 26/18/10/2 25/17/9/1 IWRWS1 TYPE3 IWRWS0 TYPE2 IWRWS0 TYPE2 IWRWS0 TYPE2 IWRWS0 TYPE2 IWRWS0 TYPE2 WM WM IWW1 TYPE1 IWW1 TYPE1 IWW1 TYPE1 IWW1 TYPE1 IWW1 TYPE1 IWW0 IWRRD1 TYPE0 IWW0 IWRRD1 TYPE0 IWW0 IWRRD1 TYPE0 IWW0 IWRRD1 TYPE0 IWW0 IWRRD1 TYPE0 SW1 BAS IWRRD0 IWRRD0 IWRRD0 IWRRD0 IWRRD0 SW0 IWRWD1 BSZ1 IWRWD1 BSZ1 IWRWD1 BSZ1 IWRWD1 BSZ1 IWRWD1 BSZ1 WR3 WR3 IWRWD0 IWRRS1 BSZ0 IWRWD0 IWRRS1 BSZ0 IWRWD0 IWRRS1 BSZ0 IWRWD0 IWRRS1 BSZ0 IWRWD0 IWRRS1 BSZ0 WR2 HW1 WR2
CS3BCR
IWRWS1 TYPE3
CS4BCR
IWRWS1 TYPE3
CS5BBCR
IWRWS1 TYPE3
CS6BBCR
IWRWS1 TYPE3
CS0WCR
WR0
CS3WCR
WR0
Rev. 6.00 Jun. 12, 2007 Page 523 of 610 REJ09B0131-0600
Section 20 List of Registers
Register Abbreviation CS3WCR (when SDRAM is in use)
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit 24/16/8/0 A3CL1 WTRC0 WW0 WR1 HW0 WW0 WR1 HW0 PCW1 THE0 WR1 HW0 PCW1 THE0 BACTV A3COL0 Module BSC
31/23/15/7 30/22/14/6 29/21/13/5 28/20/12/4 27/19/11/3 26/18/10/2 25/17/9/1 A3CL0 WTRP1 WM WM TED3 WM WM TED3 WM WTRP0 SA1 TED2 SA1 TED2 TRWL1 BAS SW1 SW1 SA0 TED1 BAS SW1 SA0 TED1 A3ROW1 WTRCD1 TRWL0 SW0 SW0 TED0 THE3 SW0 TED0 THE3 RFSH A3ROW0 WTRCD0 WW2 WR3 WW2 WR3 PCW3 THE2 WR3 PCW3 THE2 RMODE WTRC1 WW1 WR2 HW1 WW1 WR2 HW1 PCW2 THE1 WR2 HW1 PCW2 THE1 A3COL1
CS4WCR
WR0
CS5BWCR
WR0
CS5BWCR (when PCMCIA is in use)
PCW0
CS6BWCR
WR0
CS6BWCR (when PCMCIA is in use)
PCW0
SDCR

Rev. 6.00 Jun. 12, 2007 Page 524 of 610 REJ09B0131-0600
Section 20 List of Registers
Register Abbreviation RTCSR
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit 24/16/8/0 RRC0 Bit 0 Bit 0 SWR TR RR TDLA24 TDLA16 TDLA8 TDLA0 EDMAC Module BSC
31/23/15/7 30/22/14/6 29/21/13/5 28/20/12/4 27/19/11/3 26/18/10/2 25/17/9/1 CMF Bit 6 Bit 6 DE TDLA30 TDLA22 TDLA14 TDLA6 CKS2 Bit 5 Bit 5 DL1 TDLA29 TDLA21 TDLA13 TDLA5 CKS1 Bit 4 Bit 4 DL0 TDLA28 TDLA20 TDLA12 TDLA4 CKS0 Bit 3 Bit 3 TDLA27 TDLA19 TDLA11 TDLA3 RRC2 Bit 2 Bit 2 TDLA26 TDLA18 TDLA10 TDLA2 RRC1 Bit 1 Bit 1 TDLA25 TDLA17 TDLA9 TDLA1
RTCNT
Bit 7
RTCOR
Bit 7
EDMR

EDTRR

EDRRR

TDLAR
TDLA31 TDLA23 TDLA15 TDLA7
Rev. 6.00 Jun. 12, 2007 Page 525 of 610 REJ09B0131-0600
Section 20 List of Registers
Register Abbreviation RDLAR
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit 24/16/8/0 RDLA24 RDLA16 RDLA8 RDLA0 RFCOF RFOF TRO CERF RFCOFIP RFOFIP TROIP CERFIP TROCE CERFCE MFC8 MFC0 TFT8 TFT0 TFD0 RFD0 Module EDMAC
31/23/15/7 30/22/14/6 29/21/13/5 28/20/12/4 27/19/11/3 26/18/10/2 25/17/9/1 RDLA31 RDLA23 RDLA15 RDLA7 RDLA30 RDLA22 RDLA14 RDLA6 TWB ECI TWBIP ECIIP MFC14 MFC6 TFT6 RDLA29 RDLA21 RDLA13 RDLA5 TC TCIP MFC13 MFC5 TFT5 RDLA28 RDLA20 RDLA12 RDLA4 TDE RRF TDEIP RRFIP RRFCE MFC12 MFC4 TFT4 RDLA27 RDLA19 RDLA11 RDLA3 TFUF CND RTLF TFUFIP CNDIP RTLFIP CNDCE RTLFCE MFC11 MFC3 TFT3 RDLA26 RDLA18 RDLA10 RDLA2 TABT FR DLC RTSF TABTIP FRIP DLCIP RTSFIP DLCCE RTSFCE MFC10 MFC2 TFT10 TFT2 TFD2 RFD2 RDLA25 RDLA17 RDLA9 RDLA1 RABT RDE CD PRE RABTIP RDEIP CDIP PREIP CDCE PRECE MFC9 MFC1 TFT9 TFT1 TFD1 RFD1
EESR
ADE RMAF
EESIPR
ADEIP RMAFIP
TRSCER
RMAFCE
RMFCR
MFC15 MFC7
TFTR
TFT7
FDR

Rev. 6.00 Jun. 12, 2007 Page 526 of 610 REJ09B0131-0600
Section 20 List of Registers
Register Abbreviation RMCR
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit 24/16/8/0 RNC RFF0 RFD0 TIS RBWA24 RBWA16 RBWA8 RBWA0 RDFA24 RDFA16 RDFA8 RDFA0 TBRA24 TBRA16 TBRA8 TBRA0 TDFA24 TDFA16 TDFA8 TDFA0 Module EDMAC
31/23/15/7 30/22/14/6 29/21/13/5 28/20/12/4 27/19/11/3 26/18/10/2 25/17/9/1 RBWA30 RBWA22 RBWA14 RBWA6 RDFA30 RDFA22 RDFA14 RDFA6 TBRA30 TBRA22 TBRA14 TBRA6 TDFA30 TDFA22 TDFA14 TDFA6 RBWA29 RBWA21 RBWA13 RBWA5 RDFA29 RDFA21 RDFA13 RDFA5 TBRA29 TBRA21 TBRA13 TBRA5 TDFA29 TDFA21 TDFA13 TDFA5 RBWA28 RBWA20 RBWA12 RBWA4 RDFA28 RDFA20 RDFA12 RDFA4 TBRA28 TBRA20 TBRA12 TBRA4 TDFA28 TDFA20 TDFA12 TDFA4 FEC RBWA27 RBWA19 RBWA11 RBWA3 RDFA27 RDFA19 RDFA11 RDFA3 TBRA27 TBRA19 TBRA11 TBRA3 TDFA27 TDFA19 TDFA11 TDFA3 AEC RFF2 RFD2 RBWA26 RBWA18 RBWA10 RBWA2 RDFA26 RDFA18 RDFA10 RDFA2 TBRA26 TBRA18 TBRA10 TBRA2 TDFA26 TDFA18 TDFA10 TDFA2 EDH RFF1 RFD1 RBWA25 RBWA17 RBWA9 RBWA1 RDFA25 RDFA17 RDFA9 RDFA1 TBRA25 TBRA17 TBRA9 TBRA1 TDFA25 TDFA17 TDFA9 TDFA1
EDOCR

FCFTR

TRIMD

RBWAR
RBWA31 RBWA23 RBWA15 RBWA7
RDFAR
RDFA31 RDFA23 RDFA15 RDFA7
TBRAR
TBRA31 TBRA23 TBRA15 TBRA7
TDFAR
TDFA31 TDFA23 TDFA15 TDFA7
Rev. 6.00 Jun. 12, 2007 Page 527 of 610 REJ09B0131-0600
Section 20 List of Registers
Register Abbreviation ECMR
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit 24/16/8/0 TXF PRM ICD ICDIP MDC MA40 MA32 MA24 MA16 MA8 MA0 RFL8 RFL0 Module EtherC
31/23/15/7 30/22/14/6 29/21/13/5 28/20/12/4 27/19/11/3 26/18/10/2 25/17/9/1 PE MA46 MA38 MA30 MA22 MA14 MA6 RFL6 TE MA45 MA37 MA29 MA21 MA13 MA5 RFL5 PRCEF PSRTO PSRTOIP MA44 MA36 MA28 MA20 MA12 MA4 RFL4 ZPF ILB MDI MA43 MA35 MA27 MA19 MA11 MA3 RFL11 RFL3 PFR ELB LCHNG LCHNGIP MDO MA42 MA34 MA26 MA18 MA10 MA2 RFL10 RFL2 RXF MPDE DM MPD MPDIP MMD MA41 MA33 MA25 MA17 MA9 MA1 RFL9 RFL1
ECSR

ECSIPR

PIR

MAHR
MA47 MA39 MA31 MA23
MALR
MA15 MA7
RFLR
RFL7
Rev. 6.00 Jun. 12, 2007 Page 528 of 610 REJ09B0131-0600
Section 20 List of Registers
Register Abbreviation PSR
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit 24/16/8/0 LMON TROC24 TROC16 TROC8 TROC0 Module EtherC
31/23/15/7 30/22/14/6 29/21/13/5 28/20/12/4 27/19/11/3 26/18/10/2 25/17/9/1 TROC30 TROC22 TROC14 TROC6 TROC29 TROC21 TROC13 TROC5 TROC28 TROC20 TROC12 TROC4 TROC27 TROC19 TROC11 TROC3 TROC26 TROC18 TROC10 TROC2 TROC25 TROC17 TROC9 TROC1
TROCR
TROC31 TROC23 TROC15 TROC7
CDCR
COSDC31 COSDC30 COSDC29 COSDC28 COSDC27 COSDC26 COSDC25 COSDC24 COSDC23 COSDC22 COSDC21 COSDC20 COSDC19 COSDC18 COSDC17 COSDC16 COSDC15 COSDC14 COSDC13 COSDC12 COSDC11 COSDC10 COSDC9 COSDC7 COSDC6 LCC30 LCC22 LCC14 LCC6 CNDC30 CNDC22 CNDC14 CNDC6 CEFC30 CEFC22 CEFC14 CEFC6 FREC30 FREC22 FREC14 FREC6 COSDC5 LCC29 LCC21 LCC13 LCC5 CNDC29 CNDC21 CNDC13 CNDC5 CEFC29 CEFC21 CEFC13 CEFC5 FREC29 FREC21 FREC13 FREC5 COSDC4 LCC28 LCC20 LCC12 LCC4 CNDC28 CNDC20 CNDC12 CNDC4 CEFC28 CEFC20 CEFC12 CEFC4 FREC28 FREC20 FREC12 FREC4 COSDC3 LCC27 LCC19 LCC11 LCC3 CNDC27 CNDC19 CNDC11 CNDC3 CEFC27 CEFC19 CEFC11 CEFC3 FREC27 FREC19 FREC11 FREC3 COSDC2 LCC26 LCC18 LCC10 LCC2 CNDC26 CNDC18 CNDC10 CNDC2 CEFC26 CEFC18 CEFC10 CEFC2 FREC26 FREC18 FREC10 FREC2 COSDC1 LCC25 LCC17 LCC9 LCC1 CNDC25 CNDC17 CNDC9 CNDC1 CEFC25 CEFC17 CEFC9 CEFC1 FREC25 FREC17 FREC9 FREC1 COSDC8 COSDC0 LCC24 LCC16 LCC8 LCC0 CNDC24 CNDC16 CNDC8 CNDC0 CEFC24 CEFC16 CEFC8 CEFC0 FREC24 FREC16 FREC8 FREC0
LCCR
LCC 31 LCC23 LCC15 LCC7
CNDCR
CNDC31 CNDC23 CNDC15 CNDC7
CEFCR
CEFC31 CEFC23 CEFC15 CEFC7
FRECR
FREC31 FREC23 FREC15 FREC7
Rev. 6.00 Jun. 12, 2007 Page 529 of 610 REJ09B0131-0600
Section 20 List of Registers
Register Abbreviation TSFRCR
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit 24/16/8/0 TSFC24 TSFC16 TSFC8 TSFC0 TLFC24 TLFC16 TLFC8 TLFC0 RFC24 RFC16 RFC8 RFC0 MAFC24 MAFC16 MAFC8 MAFC0 IPG0 AP8 AP0 MP8 MP0 Module EtherC
31/23/15/7 30/22/14/6 29/21/13/5 28/20/12/4 27/19/11/3 26/18/10/2 25/17/9/1 TSFC31 TSFC23 TSFC15 TSFC7 TSFC30 TSFC22 TSFC14 TSFC6 TLFC30 TLFC22 TLFC14 TLFC6 RFC30 RFC22 RFC14 RFC6 MAFC30 MAFC22 MAFC14 MAFC6 AP14 AP6 MP14 MP6 TSFC29 TSFC21 TSFC13 TSFC5 TLFC29 TLFC21 TLFC13 TLFC5 RFC29 RFC21 RFC13 RFC5 MAFC29 MAFC21 MAFC13 MAFC5 AP13 AP5 MP13 MP5 TSFC28 TSFC20 TSFC12 TSFC4 TLFC28 TLFC20 TLFC12 TLFC4 RFC28 RFC20 RFC12 RFC4 MAFC28 MAFC20 MAFC12 MAFC4 IPG4 AP12 AP4 MP12 MP4 TSFC27 TSFC19 TSFC11 TSFC3 TLFC27 TLFC19 TLFC11 TLFC3 RFC27 RFC19 RFC11 RFC3 MAFC27 MAFC19 MAFC11 MAFC3 IPG3 AP11 AP3 MP11 MP3 TSFC26 TSFC18 TSFC10 TSFC2 TLFC26 TLFC18 TLFC10 TLFC2 RFC26 RFC18 RFC10 RFC2 MAFC26 MAFC18 MAFC10 MAFC2 IPG2 AP10 AP2 MP10 MP2 TSFC25 TSFC17 TSFC9 TSFC1 TLFC25 TLFC17 TLFC9 TLFC1 RFC25 RFC17 RFC9 RFC1 MAFC25 MAFC17 MAFC9 MAFC1 IPG1 AP9 AP1 MP9 MP1
TLFRCR
TLFC31 TLFC23 TLFC15 TLFC7
RFCR
RFC31 RFC23 RFC15 RFC7
MAFCR
MAFC31 MAFC23 MAFC15 MAFC7
IPGR

APR
AP15 AP7
MPR
MP15 MP7
Rev. 6.00 Jun. 12, 2007 Page 530 of 610 REJ09B0131-0600
Section 20 List of Registers
Register Abbreviation TPAUSER
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit 24/16/8/0 TPAUSE 8 Module EtherC
31/23/15/7 30/22/14/6 29/21/13/5 28/20/12/4 27/19/11/3 26/18/10/2 25/17/9/1 TPAUSE 15 TPAUSE 14 TPAUSE 13 TPAUSE 12 TPAUSE 11 TPAUSE 10 TPAUSE 9
TPAUSE7 TPAUSE6 TPAUSE5 TPAUSE4 TPAUSE3 TPAUSE2 TPAUSE1 TPAUSE0 BDRB BDB31 BDB23 BDB15 BDB7 BDMRB BDMB31 BDMB23 BDMB15 BDMB7 BRCR SCMFCA DBEB BETR BET7 BARB BAB31 BAB23 BAB15 BAB7 BAMRB BAMB31 BAMB23 BAMB15 BAMB7 BDB30 BDB22 BDB14 BDB6 BDMB30 BDMB22 BDMB14 BDMB6 SCMFCB PCBB BET6 BAB30 BAB22 BAB14 BAB6 BAMB30 BAMB22 BAMB14 BAMB6 BDB29 BDB21 BDB13 BDB5 BDMB29 BDMB21 BDMB13 BDMB5 SCMFDA BET5 BAB29 BAB21 BAB13 BAB5 BAMB29 BAMB21 BAMB13 BAMB5 BDB28 BDB20 BDB12 BDB4 BDMB28 BDMB20 BDMB12 BDMB4 SCMFDB BET4 BAB28 BAB20 BAB12 BAB4 BAMB28 BAMB20 BAMB12 BAMB4 BDB27 BDB19 BDB11 BDB3 BDMB27 BDMB19 BDMB11 BDMB3 PCTE SEQ BET11 BET3 BAB27 BAB19 BAB11 BAB3 BAMB27 BAMB19 BAMB11 BAMB3 BDB26 BDB18 BDB10 BDB2 BDMB26 BDMB18 BDMB10 BDMB2 PCBA BET10 BET2 BAB26 BAB18 BAB10 BAB2 BAMB26 BAMB18 BAMB10 BAMB2 BDB25 BDB17 BDB9 BDB1 BDMB25 BDMB17 BDMB9 BDMB1 BET9 BET1 BAB25 BAB17 BAB9 BAB1 BAMB25 BAMB17 BAMB9 BAMB1 BDB24 BDB16 BDB8 BDB0 BDMB24 BDMB16 BDMB8 BDMB0 ETBE BET8 BET0 BAB24 BAB16 BAB8 BAB0 BAMB24 BAMB16 BAMB8 BAMB0 UBC
Rev. 6.00 Jun. 12, 2007 Page 531 of 610 REJ09B0131-0600
Section 20 List of Registers
Register Abbreviation BBRB
Bit
Bit
Bit
Bit
Bit
Bit
Bit
Bit 24/16/8/0 SZB0 BSA24 BSA16 BSA8 BSA0 BAA24 BAA16 BAA8 BAA0 BAMA24 BAMA16 BAMA8 BAMA0 SZA0 BDA24 BDA16 BDA8 BDA0 CE Cache Module UBC
31/23/15/7 30/22/14/6 29/21/13/5 28/20/12/4 27/19/11/3 26/18/10/2 25/17/9/1 CDB1 CDB0 BSA22 BSA14 BSA6 BAA30 BAA22 BAA14 BAA6 BAMA30 BAMA22 BAMA14 BAMA6 CDA0 BDA22 BDA14 BDA6 IDB1 BSA21 BSA13 BSA5 BAA29 BAA21 BAA13 BAA5 BAMA29 BAMA21 BAMA13 BAMA5 IDA1 BDA21 BDA13 BDA5 IDB0 BSA20 BSA12 BSA4 BAA28 BAA20 BAA12 BAA4 BAMA28 BAMA20 BAMA12 BAMA4 IDA0 BDA20 BDA12 BDA4 RWB1 BSA27 BSA19 BSA11 BSA3 BAA27 BAA19 BAA11 BAA3 BAMA27 BAMA19 BAMA11 BAMA3 RWA1 BDA27 BDA19 BDA11 BDA3 CF RWB0 BSA26 BSA18 BSA10 BSA2 BAA26 BAA18 BAA10 BAA2 BAMA26 BAMA18 BAMA10 BAMA2 RWA0 BDA26 BDA18 BDA10 BDA2 CB SZB1 BSA25 BSA17 BSA9 BSA1 BAA25 BAA17 BAA9 BAA1 BAMA25 BAMA17 BAMA9 BAMA1 SZA1 BDA25 BDA17 BDA9 BDA1 WT
BRSR
SVF BSA23 BSA15 BSA7
BARA
BAA31 BAA23 BAA15 BAA7
BAMRA
BAMA31 BAMA23 BAMA15 BAMA7
BBRA
CDA1
BRDR
DVF BDA23 BDA15 BDA7
CCR1

Note:
*
Supported only by the SH7618A.
Rev. 6.00 Jun. 12, 2007 Page 532 of 610 REJ09B0131-0600
Section 20 List of Registers
20.3
Module Cache I/O
Register States in Each Processing State
Register Abbreviation CCR3*
4
Power-On Address H'F80000B4 H'F8050000 H'F8050004 H'F8050008 H'F805000A H'F8050012 H'F8050016 H'F805001C H'F805001E H'F8050020 H'F8050022 H'F8050024 H'F8050026 H'F805002A H'F805002C H'F805002E H'F8050032 H'F8050036 H'F805003E H'F8050040 H'F8050042 H'F8050044 H'F8050046 H'F8050048 H'F805004A H'F805004C H'F805004E Reset Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized
Software Standby Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained
Module Standby Retained * *
3
Sleep Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained
PADRH PAIORH PACRH1 PACRH2 PBDRL PBIORL PBCRL1 PBCRL2 PCDRH PCDRL PCIORH PCIORL PCCRH2 PCCRL1 PCCRL2 PDDRL PDIORL PDCRL2 PEDRH PEDRL PEIORH PEIORL PECRH1 PECRH2 PECRL1 PECRL2
3
*3 * * * * * * *
3
3
3
3
3
3
3
*3 * * * * * * *
3
3
3
3
3
3
3
*3 * * * * * * *
3
3
3
3
3
3
3
Rev. 6.00 Jun. 12, 2007 Page 533 of 610 REJ09B0131-0600
Section 20 List of Registers
Register Module INTC Abbreviation IPRC IPRD IPRE Power-down mode STBCR3 STBCR4 CPG H-UDI MCLKCR SDIR SDID INTC ICR0 IRQCR IRQSR IPRA IPRB CPG Power-down mode WDT FRQCR STBCR WTCNT WTCSR Power-down mode SCIF_0 STBCR2 SCSMR_0 SCBRR_0 SCSCR_0 SCFTDR_0 SCFSR_0 SCFRDR_0 SCFCR_0 SCFDR_0 SCSPTR_0 SCLSR_0 SCIF_1 SCSMR_1 SCBRR_1 SCSCR_1 Address H'F8080000 H'F8080002 H'F8080004 H'F80A0000 H'F80A0004 H'F80A000C H'F8100200 H'F8100214 H'F8140000 H'F8140002 H'F8140004 H'F8140006 H'F8140008 H'F815FF80 H'F815FF82 H'F815FF84 H'F815FF86 H'F815FF88 H'F8400000 H'F8400004 H'F8400008 H'F840000C H'F8400010 H'F8400014 H'F8400018 H'F840001C H'F8400020 H'F8400024 H'F8410000 H'F8410004 H'F8410008
Power-On Reset Initialized Initialized Initialized Initialized Initialized Initialized Retained Retained Initialized* Initialized Initialized* Initialized Initialized Initialized* Initialized Initialized* Initialized* Initialized Initialized Initialized Initialized Undefined Initialized Undefined Initialized Initialized Initialized*1 Initialized Initialized Initialized Initialized
2 2 1 1
Software Standby Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained
Module Standby *
3
Sleep Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained
*3 * *
3
3
* *
3
3
Retained Retained *
3
*3 *
3
* * *
3
3
3
* * *
3
3
2
3
*3 Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained
Rev. 6.00 Jun. 12, 2007 Page 534 of 610 REJ09B0131-0600
Section 20 List of Registers
Register Module SCIF_1 Abbreviation SCFTDR_1 SCFSR_1 SCFRDR_1 SCFCR_1 SCFDR_1 SCSPTR_1 SCLSR_1 SCIF_2 SCSMR_2 SCBRR_2 SCSCR_2 SCFTDR_2 SCFSR_2 SCFRDR_2 SCFCR_2 SCFDR_2 SCSPTR_2 SCLSR_2 CMT CMSTR CMCSR_0 CMCNT_0 CMCOR_0 CMCSR_1 CMCNT_1 CMCOR_1 HIF HIFIDX HIFGSR HIFSCR HIFMCR Address H'F841000C H'F8410010 H'F8410014 H'F8410018 H'F841001C H'F8410020 H'F8410024 H'F8420000 H'F8420004 H'F8420008 H'F842000C H'F8420010 H'F8420014 H'F8420018 H'F842001C H'F8420020 H'F8420024 H'F84A0070 H'F84A0072 H'F84A0074 H'F84A0076 H'F84A0078 H'F84A007A H'F84A007C H'F84D0000 H'F84D0004 H'F84D0008 H'F84D000C
Power-On Reset Undefined Initialized Undefined Initialized Initialized Initialized* Initialized Initialized Initialized Initialized Undefined Initialized Undefined Initialized Initialized Initialized* Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized* Initialized
1 1 1
Software Standby Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Initialized Initialized Initialized Initialized Initialized Initialized Initialized Retained Retained Retained Retained
Module Standby Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Sleep Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained
Rev. 6.00 Jun. 12, 2007 Page 535 of 610 REJ09B0131-0600
Section 20 List of Registers
Register Module HIF Abbreviation HIFIICR HIFEICR HIFADR HIFDATA HIFDTR HIFBICR HIFBCR BSC CMNCR CS0BCR CS3BCR CS4BCR CS5BBCR CS6BBCR CS0WCR CS3WCR CS3WCR (SDRAM in use) CS4WCR CS5BWCR CS5BWCR (PCMCIA in use) CS6BWCR CS6BWCR (PCMCIA in use) SDCR RTCSR RTCNT RTCOR E-DMAC EDMR EDTRR EDRRR H'F8FD0044 H'F8FD0048 H'F8FD004C H'F8FD0050 H'FB000000 H'FB000004 H'FB000008 H'F8FD0040 H'F8FD0040 H'F8FD0030 H'F8FD0038 H'F8FD0038 Address H'F84D0010 H'F84D0014 H'F84D0018 H'F84D001C H'F84D0020 H'F84D0024 H'F84D0040 H'F8FD0000 H'F8FD0004 H'F8FD000C H'F8FD0010 H'F8FD0018 H'F8FD0020 H'F8FD0024 H'F8FD002C H'F8FD002C
Power-On Reset Initialized Initialized Initialized Initialized Initialized Initialized Initialized* Initialized* Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized
1
Software Standby Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained
Module Standby Retained Retained Retained Retained Retained Retained Retained *
3
Sleep Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained
1
*
3
*3 * * * * * *
3
3
3
3
3
3
Initialized Initialized Initialized
Retained Retained Retained
*3 * *
3
Retained Retained Retained
3
Initialized Initialized
Retained Retained
*3 *
3
Retained Retained
Initialized Initialized Initialized Initialized Initialized Initialized Initialized
Retained Retained Retained Retained Retained Retained Retained
*3 * * *
3
Retained Retained Retained Retained Retained Retained Retained
3
3
Retained Retained Retained
Rev. 6.00 Jun. 12, 2007 Page 536 of 610 REJ09B0131-0600
Section 20 List of Registers
Register Module E-DMAC Abbreviation TDLAR RDLAR EESR EESIPR TRSCER RMFCR TFTR FDR RMCR EDOCR FCFTR TRIMD RBWAR RDFAR TBRAR TDFAR EtherC ECMR ECSR ECSIPR PIR MAHR MALR RFLR PSR TROCR CDCR LCCR CNDCR CEFCR FRECR TSFRCR Address H'FB00000C H'FB000010 H'FB000014 H'FB000018 H'FB00001C H'FB000020 H'FB000024 H'FB000028 H'FB00002C H'FB000030 H'FB000034 H'FB00003C H'FB000040 H'FB000044 H'FB00004C H'FB000050 H'FB000160 H'FB000164 H'FB000168 H'FB00016C H'FB000170 H'FB000174 H'FB000178 H'FB00017C H'FB000180 H'FB000184 H'FB000188 H'FB00018C H'FB000194 H'FB000198 H'FB00019C
Power-On Reset Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized* Initialized Initialized Initialized Initialized* Initialized Initialized Initialized Initialized Initialized Initialized Initialized
1 1
Software Standby Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained
Module Standby Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Sleep Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained
Rev. 6.00 Jun. 12, 2007 Page 537 of 610 REJ09B0131-0600
Section 20 List of Registers
Register Module EtherC Abbreviation TLFRCR RFCR MAFCR IPGR APR MPR TPAUSER UBC BDRB BDMRB BRCR BETR BARB BAMRB BBRB BRSR BARA BAMRA BBRA BRDR Cache CCR1 Address H'FB0001A0 H'FB0001A4 H'FB0001A8 H'FB0001B4 H'FB0001B8 H'FB0001BC H'FB0001C4 H'FFFFFF90 H'FFFFFF94 H'FFFFFF98 H'FFFFFF9C H'FFFFFFA0 H'FFFFFFA4 H'FFFFFFA8 H'FFFFFFAC H'FFFFFFB0 H'FFFFFFB4 H'FFFFFFB8 H'FFFFFFBC H'FFFFFFEC
Power-On Reset Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized Initialized* Initialized Initialized Initialized Initialized*1 Initialized
1
Software Standby Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained
Module Standby Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Sleep Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained Retained
Notes: 1. 2. 3. 4.
Some bits are not initialized. Not initialized by a power-on reset caused by the WDT. This module does not enter the module standby mode. Supported only by the SH7618A.
Rev. 6.00 Jun. 12, 2007 Page 538 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Section 21 Electrical Characteristics
21.1 Absolute Maximum Ratings
Table 21.1 shows the absolute maximum ratings. Table 21.1 Absolute Maximum Ratings
Item Power supply voltage (I/O) Power supply voltage (internal) Symbol VCCQ VCC, VCC (PLL1), VCC (PLL2) Vin Regular specifications Wide-range specifications Storage temperature Tstg Topr Value -0.3 to +4.2 -0.3 to +2.5 Unit V V
Input voltage Operating temperature
-0.3 to VCCQ + 0.3 -20 to +75 -40 to +85 -55 to +125
V C C C
Caution: Permanent damage to the LSI may result if absolute maximum ratings are exceeded.
Rev. 6.00 Jun. 12, 2007 Page 539 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
21.2
Power-On and Power-Off Order
* Order of turning on 1.5-V system power (VCC (main), VCC (sub), VCC (PLL1), and VCC (PLL2)) and 3.3-V system power (VCCQ) First turn on the 3.3-V system power, then turn on the 1.5-V system power within 1 ms. This time should be as short as possible. The system design must ensure that the states of pins or undefined period of an internal state do not cause erroneous system operation. Until voltage is applied to all power supplies and a low level is input to the RES pin, internal circuits remain unsettled, and so pin states are also undefined. The system design must ensure that these undefined states do not cause erroneous system operation. Waveforms at power-on are shown in the following figure.
VccQ : 3.3-V system power VccQ (min.) power Vcc : 1.5-V system power tPWU Vcc (min.) Vcc/2 GND tUNC Pin states undefined RES Input low level in advance Normal operation period
Other pins * Pin states undefined Note: * Except power/GND and clock related pins Power-on reset state
Table 21.2 Recommended Timing at Power-On
Item Time difference between turning on VCCQ and VCC Time over which the internal state is undefined Note: * Symbol tPWU tUNC Maximum Value 1 100 Unit ms ms
The values shown in table 21.2 are recommended values, so they represent guidelines rather than strict requirements.
Rev. 6.00 Jun. 12, 2007 Page 540 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
The time over which the internal state is undefined means the time taken to reach VCC (min.). The pin states become settled when VCCQ reached the VCCQ (min.). The timing when a power-on reset (RES) is normally accepted is after VCC reaches VCC (min.) and oscillation becomes stable (when using the on-chip oscillator). Ensure that the time over which the internal state is undefined is less than or equal to 100 ms. * Power-off order In the reverse order of power-on, first turn off the 1.5-V system power, then turn off the 3.3-V system power within 10 ms. This time should be as short as possible. The system design must ensure that the states of pins or undefined period of an internal state do not cause erroneous system operation. Pin states are undefined while only the 1.5-V system power is turned off. The system design must ensure that these undefined states do not cause erroneous system operation.
VccQ : 3.3-V system power
Vcc : 1.5-V system power
tPWD Vcc/2
GND Normal operation period Operation stopped
Table 21.3 Recommended Timing in Power-Off
Item Time difference between turning off VCCQ and VCC Note: * Symbol tPWD Maximum Value 10 Unit ms
The table shown above is recommended values, so they represent guidelines rather than strict requirements.
Rev. 6.00 Jun. 12, 2007 Page 541 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
21.3
DC Characteristics
Tables 21.4 and 21.5 show the DC characteristics. Table 21.4 DC Characteristics (1) Conditions: Ta = -20C to +75C (regular specifications), Ta = -40C to +85C (wide-range specifications)
Item Current consumption Normal operation Symbol ICC ICCQ Standby mode Istby Min. Typ. 100 30 500 Max. Unit 140 50 700 mA mA A Test Conditions VCC = 1.5 V VCCQ = 3.3 V I = 100 MHz B = 50 MHz Ta = 25C VCC = 1.5 V VCCQ = 3.3 V VCC = 1.5 V VCCQ = 3.3 V B = 50 MHz Vin = 0.5 to VCCQ - 0.5 V Vin = 0.5 to VCCQ - 0.5 V
Sleep mode
Isleep IsleepQ

40 30
60 50 1.0 1.0 10
mA mA A A pF
Input leakage current Tri-state leakage current Input capacitance
All pins
| Iin |
I/O pins, all output | ISTI | pins (off state) All pins C
Rev. 6.00 Jun. 12, 2007 Page 542 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Table 21.4 DC Characteristics (2) Conditions: Ta = -20C to +75C (regular specifications), Ta = -40C to +85C (wide-range specifications)
Item Power supply Symbol VCCQ Min. 3.0 Typ. 3.3 1.5 Max. 3.6 1.6 Test Unit Conditions V
1.4 VCC, VCC (PLL1), VCC (PLL2) RES, NMI, IRQ7 VIH to IRQ0, MD5, MD3 to MD0, ASEMD, TESTMD, HIFMD, TRST EXTAL Other input pins RES, NMI, IRQ7 VIL to IRQ0, MD5, MD3 to MD0, ASEMD, TESTMD, HIFMD, TRST EXTAL Other input pins
Input high voltage
VCCQ x 0.9
VCCQ + 0.3 V
VCCQ - 0.3 2.0 -0.3
VCCQ + 0.3 VCCQ + 0.3 VCCQ x 0.1 V
Input low voltage
-0.3 -0.3 VOH 2.4 2.0

VCCQ x 0.2 VCCQ x 0.2 0.55 V V VCCQ = 3.0 V IOH = -200 A VCCQ = 3.0 V IOH = -2 mA VCCQ = 3.6 V IOL = 2.0 mA
Output high All output pins voltage
Output low voltage
All output pins
VOL
Notes: 1. The VCC and VSS pins must be connected to the VCC and VSS. 2. Current consumption values are for VIH min. = VCCQ - 0.5 V and VIL max. = 0.5 V with all output pins unloaded.
Rev. 6.00 Jun. 12, 2007 Page 543 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Table 21.5 Permissible Output Currents Conditions: VCCQ = 3.0 V to 3.6 V, VCC = 1.4 V to 1.6 V, Ta = -20C to +75C (regular specifications), Ta = -40C to +85C (wide-range specifications)
Item Permissible output low current (per pin) Permissible output low current (total) Permissible output high current (per pin) Permissible output high current (total) Caution: Symbol IOL IOL -IOH -IOH Min. Typ. Max. 2.0 120 2.0 40 Unit mA mA mA mA
To protect the LSI's reliability, do not exceed the output current values in table 21.5.
21.4
AC Characteristics
Signals input to this LSI are basically handled as signals synchronized with the clock. Unless otherwise noted, setup and hold times for individual signals must be followed. Table 21.6 Maximum Operating Frequency Conditions: VCCQ = 3.0 V to 3.6 V, VCC = 1.4 V to 1.6 V, Ta = -20C to +75C (regular specifications), Ta = -40C to +85C (wide-range specifications)
Item Operating frequency CPU, cache (I) External bus (B) On-chip peripheral module (P) Symbol f Min. 20 20 5 Typ. Max. 100 50 50 Unit MHz Test Conditions
Rev. 6.00 Jun. 12, 2007 Page 544 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
21.4.1
Clock Timing
Table 21.7 Clock Timing Conditions: VCCQ = 3.0 V to 3.6 V, VCC = 1.4 V to 1.6 V, Ta = -20C to +75C (regular specifications), Ta = -40C to +85C (wide-range specifications), External bus operating frequency (Max.) = 50 MHz,
Item EXTAL clock input frequency EXTAL clock input cycle time EXTAL clock input low pulse width EXTAL clock input high pulse width EXTAL clock rising time EXTAL clock falling time CKIO clock output frequency CKIO clock output cycle time CKIO clock low pulse width CKIO clock high pulse width CKIO clock rising time CKIO clock falling time CK_PHY clock low pulse width Symbol fEX tExcyc tEXL tEXH tExr tExf fOP tcyc tCKOL tCKOH tCKOr tCKOf tCKPHYL tCKPHYr tCKPHYf tOSC1 tRESS tRESW tOSC2 Min. 10 40 10 10 20 20 5 5 12 12 10 25 20 10 Max. 25 100 4 4 50 50 5 5 6 6 Unit. MHz ns ns ns ns ns MHz ns ns ns ns ns ns ns ns ns ms ns tbcyc* ms Figure 21.4 Figure 21.3 Figures 21.3 and 21.4 Figure 21.2 Reference Figures Figure 21.1
CK_PHY clock high pulse width tCKPHYH CK_PHY clock rising time CK_PHY clock falling time Oscillation settling time (power-on) RES setup time RES assert time Oscillation settling time 1 (leaving standby mode)
Rev. 6.00 Jun. 12, 2007 Page 545 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Item Oscillation settling time 2 (leaving standby mode) PLL synchronize settling time Note: *
Symbol tOSC3 tPLL
Min.
Max. 10 100
Unit. ms s
Reference Figures Figure 21.5 Figure 21.6
tbcyc indicates the period of the external bus clock (B).
tEXcyc tEXH tEXL VIH 1/2 Vcc tEXr
EXTAL* (input) 1/2 Vcc
VIH
VIH VIL tEXf VIL
Note: * When the clock is input to the EXTAL pin
Figure 21.1 External Clock Input Timing
tcyc tCKOH tCKOL
CKIO (output)
1/2 Vcc
VOH
VOH VOL tCKOf
VOH VOL
1/2 Vcc tCKOr
40 ns tCKPHYH tCKPHYL
CK_PHY
1/2 Vcc
VOH
VOH VOL
VOH VOL
1/2 Vcc tCKPHYr
tCKPHYf
Note: The CK_PHY output timing is the timing when its frequency is set to 25 MHz.
Figure 21.2 CKIO and CK_PHY Clock Output Timings
Rev. 6.00 Jun. 12, 2007 Page 546 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Oscillation settled CKIO, internal clock Vcc Vcc min. tOSC1 RES
tRESW
tRESS
Note: Oscillation settling time when the internal oscillator is in use
Figure 21.3 Oscillation Settling Timing after Power-On
Standby mode CKIO, internal clock tRESW Oscillation settled
tOSC2 RES
Note: Oscillation settling time when the internal oscillator is in use
Figure 21.4 Oscillation Settling Timing after Standby Mode (By Reset)
Standby mode CKIO, internal clock tOSC3 Oscillation settled
NMI
Note: Oscillation settling time when the internal oscillator is in use
Figure 21.5 Oscillation Settling Timing after Standby Mode (By NMI or IRQ)
Rev. 6.00 Jun. 12, 2007 Page 547 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Reset or NMI interrupt request Input clock settled
Standby mode
Input clock settled
EXTAL input PLL synchronization PLL output, CKIO output
tPLL
PLL synchronization
Internal clock
Note: PLL oscillation settling time when the clock is input to the EXTAL pin
Figure 21.6 PLL Synchronize Settling Timing By Reset or NMI
Rev. 6.00 Jun. 12, 2007 Page 548 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
21.4.2
Control Signal Timing
Table 21.8 Control Signal Timing Conditions: VCCQ = 3.0 V to 3.6 V, VCC = 1.4 V to 1.6 V, Ta = -20C to +75C (regular specifications), Ta = -40C to +85C (wide-range specifications)
Item RES pulse width RES setup time* RES hold time NMI setup time* NMI hold time IRQ7 to IRQ0 setup time* IRQ7 to IRQ0 hold time Bus tri-state delay time 1 Bus tri-state delay time 2 Bus buffer on time 1 Bus buffer on time 2
1 1 1
Symbol tRESW tRESS tRESH tNMIS tNMIH tIRQS tIRQH tBOFF1 tBOFF2 tBON1 tBON2
Min. 20*2 25 15 12 10 12 10
Max. 20 20 20 20
Unit tbcyc*3 ns ns ns ns ns ns ns ns ns ns
Reference Figures Figures 21.7 and 21.8
Figure 21.8
Figure 21.9
Notes: 1. The RES, NMI, and IRQ7 to IRQ0 signals are asynchronous signals. When the setup time is satisfied, a signal change is detected at the rising edge of the clock signal. When the setup time is not satisfied, a signal change may be delayed to the next rising edge. 2. In standby mode, tRESW = tOSC2 (10 ms). When changing the clock multiplication, tRESW = tPLL1 (100 s). 3. tbcyc indicates the period of the external bus clock (B).
CKIO tRESS tRESW RES tRESS
Figure 21.7 Reset Input Timing
Rev. 6.00 Jun. 12, 2007 Page 549 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
CKIO tRESH RES tNMIH NMI tIRQH IRQ7 to IRQ0 VIL tRESS VIH VIL tNMIS VIH VIL tIRQS VIH
Figure 21.8 Interrupt Input Timing
Normal mode
CKIO tBOFF2 RD, RD/WR, RAS, CAS, CSn, WEn, BS, CKE tBOFF1 A25 to A0, D15 to D0 tBON1 tBON2
Standby mode
Normal mode
Figure 21.9 Pin Drive Timing in Standby Mode
Rev. 6.00 Jun. 12, 2007 Page 550 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
21.4.3
Bus Timing
Table 21.9 Bus Timing Conditions: Clock mode = 1/2/5/6, VCCQ = 3.0 V to 3.6 V, VCC = 1.4 V to 1.6 V, Ta = -20C to +75C (regular specifications), Ta = -40C to +85C (wide-range specifications)
Item Address delay time 1 Address setup time Address hold time BS delay time CS delay time 1 Read write delay time Write strobe delay time Read strobe time Read data setup time 1 Read data setup time 2 Read data hold time 1 Read data hold time 2 Symbol Min. tAD1 tAS tAH tBSD tCSD1 tRWD tRWD2 tRSD tRDS1 tRDS2 tRDH1 tRDH2 1 3 3 0 1 1 1/2 x tbcyc Max. 15 14 14 14 14 Unit ns ns ns ns ns ns ns Reference Figures Figures 21.10 to 21.36 Figures 21.10 to 21.13 Figures 21.10 to 21.13 Figures 21.10 to 21.29 and 21.33 to 21.36 Figures 21.10 to 21.36 Figures 21.10 to 21.36 Figure 21.15 Figures 21.10 to 21.15, 21.33, and 21.34 Figures 21.10 to 21.15 and 21.33 to 21.36 Figures 21.16 to 21.19 and 21.24 to 21.26 Figures 21.10 to 21.15 and 21.33 to 21.36 Figures 21.16 to 21.19 and 21.24 to 21.26 Figures 21.10 to 21.14, 21.33, and 21.34 Figure 21.15 Figures 21.10 to 21.15 and 21.33 to 21.36 Figures 21.20 to 21.23 and 21.27 to 21.29 Figures 21.10 to 21.15 and 21.33 to 21.36
1/2 x tbcyc + 13 ns ns ns ns ns
1/2 x tbcyc + 10 12 0 2 1/2 x tbcyc 2
Write enable delay time 1 tWED1 Write enable delay time 2 tWED2 Write data delay time 1 Write data delay time 2 Write data hold time 1 tWDD1 tWDD2 tWDH1
1/2 x tbcyc + 13 ns 13 18 17 ns ns ns ns
Rev. 6.00 Jun. 12, 2007 Page 551 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Item Write data hold time 2 Write data hold time 3 WAIT setup time WAIT hold time RAS delay time CAS delay time DQM delay time CKE delay time ICIORD delay time ICIOWR delay time IOIS16 setup time IOIS16 hold time
Symbol tWDH2 tWDH3 tWTS tWTH tRASD tCASD tDQMD tCKED tICRSD tICWSD tIO16S tIO16H
Min. 2 0
Max.
Unit ns ns ns ns ns ns ns ns
Reference Figures Figures 21.20 to 21.23 and 21.27 to 21.29 Figures 21.10 to 21.13 Figures 21.12 to 21.15, 21.34, and 21.36 Figures 21.12 to 21.15, 21.34, and 21.36 Figures 21.16 to 21.27 and 21.29 to 21.32 Figures 21.16 to 21.32 Figures 21.16 to 21.29 Figure 21.31 Figures 21.35 and 21.36 Figures 21.35 and 21.36 Figure 21.36 Figure 21.36
1/2 x tbcyc + 11 1/2 x tbcyc + 10 1 1 1 1/2 x tbcyc 1/2 x tbcyc 15 15 15 14
1/2 x tbcyc + 15 ns 1/2 x tbcyc + 15 ns ns ns
1/2 x tbcyc + 11 1/2 x tbcyc + 10
Rev. 6.00 Jun. 12, 2007 Page 552 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
21.4.4
Basic Timing
T1 T2
CKIO
tAD1 tAD1
A25 to A0
tAS tCSD1 tCSD1
CSn
tRWD
tRWD
RD/WR
tRSD tRSD tAH tRDH1 tRDS1
RD Read D15 to D0
tWED1
tWED1
tAH tWDH3
WEn Write D15 to D0
tWDD1
tWDH1
tBSD
tBSD
BS
Figure 21.10 Basic Bus Timing: No Wait Cycle
Rev. 6.00 Jun. 12, 2007 Page 553 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
T1
Tw
T2
CKIO
tAD1 tAD1
A25 to A0
tAS tCSD1 tCSD1
CSn
tRWD
tRWD
RD/WR
tRSD tRSD tAH
RD Read D15 to D0
tRDS1
tRDH1
tWED1
tWED1
tAH
WEn Write D15 to D0
tWDD1
tWDH3 tWDH1
tBSD
tBSD
BS
tWTH tWTS
WAIT
Figure 21.11 Basic Bus Timing: One Software Wait Cycle
Rev. 6.00 Jun. 12, 2007 Page 554 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
T1
TwX
T2
CKIO
tAD1 tAD1
A25 to A0
tAS tCSD1 tCSD1
CSn
tRWD tRWD
RD/WR
tRSD tRSD tAH
RD Read D15 to D0
tRDS1
tRDH1
tWED1
tWED1
tAH
WEn Write D15 to D0
tWDD1
tWDH3 tWDH1
tBSD
tBSD
BS
tWTH tWTS
tWTH tWTS
WAIT
Figure 21.12 Basic Bus Timing: One External Wait Cycle
Rev. 6.00 Jun. 12, 2007 Page 555 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
T1
Tw
T2
Taw
T1
Tw
T2
Taw
CKIO
tAD1 tAD1 tAD1 tAD1
A25 to A0
tAS tCSD1 tCSD1 tAS tCSD1 tCSD1
CSn
tRWD tRWD tRWD tRWD
RD/WR
tRSD tRSD tAH tRSD tRSD tAH
RD Read D15 to D0
tWED1 tWED1
tRDH1 tRDS1 tRDS1
tRDH1
tAH
tWED1
tWED1
tAH
WEn
tWDH3 tWDH3
Write D15 to D0
tWDD1
tWDH1
tWDD1
tWDH1
tBSD
tBSD
tBSD
tBSD
BS
tWTH tWTS tWTH tWTS
WAIT
Figure 21.13 Basic Bus Timing: One Software Wait Cycle, External Wait Enabled (WM Bit = 0), No Idle Cycle
Rev. 6.00 Jun. 12, 2007 Page 556 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Th CKIO tAD1 An tCSD1 CSn
T1
Twx
T2
Tf
tAD1
tCSD1
tWED1 WEn tRWD RD/WR tRSD
tWED1
tRWD
tRSD
Read
RD tRDS1 D15 to D0 tRWD RD/WR
tRDH1
tRWD
Write
D15 to D0
tWDD1
tWDH1
tBSD BS
tBSD
tWTH WAIT tWTS
tWTH
tWTS
Figure 21.14 Byte Control SRAM Timing: SW = 1 Cycle, HW = 1 Cycle, One Asynchronous External Wait Cycle, CSnWCR.BAS = 0 (UB-/LB-Controlled Write Cycle)
Rev. 6.00 Jun. 12, 2007 Page 557 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Th
T1
Twx
T2
Tf
CKIO
tAD1 tAD1
An
tCSD1 tCSD1
CSn
tWED2 tWED2
WEn
tRWD tRWD
RD/WR
tRSD tRSD
Read
RD
tRDS1
tRDH1
D15 to D0
tRWD
tRWD2
tRWD2
tRWD
RD/WR
Write
D15 to D0
tWDD1
tWDH1
tBSD
tBSD
BS
tWTH tWTH
WAIT
tWTS tWTS
Figure 21.15 Byte Control SRAM Timing: SW = 1 Cycle, HW = 1 Cycle, One Asynchronous External Wait Cycle, CSnWCR.BAS = 1 (WE-Controlled Write Cycle)
Rev. 6.00 Jun. 12, 2007 Page 558 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
21.4.5
Synchronous DRAM Timing
Tr Tc1 Tcw Td1 Tde
CKIO
tAD1 tAD1 Row address Column address tAD1
A25 to A0
tAD1
tAD1
tAD1
A11*
Read A command
tCSD1
tCSD1
CSn
tRWD tRWD
RD/WR
tRASD tRASD
RAS
tCASD tCASD
CAS
tDQMD tDQMD
DQMxx
tRDS2 tRDH2
D15 to D0
tBSD tBSD
BS
(High)
CKE
Note: * An address pin connected to pin A10 of SDRAM
Figure 21.16 Synchronous DRAM Single Read Bus Cycle (Auto-Precharge, CAS Latency = 2, WTRCD = 0 Cycle, WTRP = 0 Cycle)
Rev. 6.00 Jun. 12, 2007 Page 559 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Tr
Trw
Tc1
Tcw
Td1
Tde
Tap
CKIO
tAD1 tAD1 Row address Column address tAD1
A25 to A0
tAD1
tAD1
tAD1
A11*
Read A command
tCSD1
tCSD1
CSn
tRWD tRWD
RD/WR
tRASD tRASD
RAS
tCASD tCASD
CAS
tDQMD tDQMD
DQMxx
tRDS2 tRDH2
D15 to D0
tBSD tBSD
BS
(High)
CKE
Note: * An address pin connected to pin A10 of SDRAM
Figure 21.17 Synchronous DRAM Single Read Bus Cycle (Auto-Precharge, CAS Latency = 2, WTRCD = 1 Cycle, WTRP = 1 Cycle)
Rev. 6.00 Jun. 12, 2007 Page 560 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Td1 Tr Tc1 Tc2 Tc3
Td2 Tc4
Td3
Td4 Tde
CKIO
tAD1 tAD1 Column address tAD1 Column address tAD1 Column address tAD1 Column address tAD1
A25 to A0
Row address
tAD1
tAD1 Read command
tAD1
tAD1
A11*
tCSD1
Read A command
tCSD1
CSn
tRWD tRWD
RD/WR
tRASD tRASD
RAS
tCASD tCASD
CAS
tDQMD tDQMD
DQMxx
tRDS2 tRDH2 tRDS2 tRDH2
D15 to D0
tBSD tBSD
BS
(High)
CKE
Note: * An address pin connected to pin A10 of SDRAM
Figure 21.18 Synchronous DRAM Burst Read Bus Cycle (Single Read x 4) (Auto-Precharge, CAS Latency = 2, WTRCD = 0 Cycle, WTRP = 1 Cycle)
Rev. 6.00 Jun. 12, 2007 Page 561 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Td1 Tr Trw Tc1 Tc2 Tc3
Td2 Tc4
Td3
Td4 Tde
CKIO
tAD1 tAD1 Row address Column address tAD1 Column address tAD1 Column address tAD1 Column address tAD1
A25 to A0
tAD1
tAD1 Read command
tAD1
tAD1
A11*
Read A command
tCSD1
tCSD1
CSn
tRWD tRWD
RD/WR
tRASD tRASD
RAS
tCASD tCASD
CAS
tDQMD tDQMD
DQMxx
tRDS2 tRDH2 tRDS2 tRDH2
D15 to D0
tBSD tBSD
BS
(High)
CKE
Note: * An address pin connected to pin A10 of SDRAM
Figure 21.19 Synchronous DRAM Burst Read Bus Cycle (Single Read x 4) (Auto-Precharge, CAS Latency = 2, WTRCD = 1 Cycle, WTRP = 0 Cycle)
Rev. 6.00 Jun. 12, 2007 Page 562 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Tr
Tc1
Trwl
CKIO
tAD1 tAD1 tAD1
A25 to A0
Row address Column address
tAD1
tAD1
tAD1
A11*
Write A command
tCSD1
tCSD1
CSn
tRWD tRWD tRWD
RD/WR
tRASD tRASD
RAS
tCASD tCASD
CAS
tDQMD tDQMD
DQMxx
tWDD2 tWDH2
D15 to D0
tBSD tBSD
BS
(High)
CKE
Note: * An address pin connected to pin A10 of SDRAM
Figure 21.20 Synchronous DRAM Single Write Bus Cycle (Auto-Precharge, TRWL = 1 Cycle)
Rev. 6.00 Jun. 12, 2007 Page 563 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Tr
Trw
Trw
Tc1
Trwl
CKIO
tAD1 tAD1 Row address tAD1
A25 to A0
Column address
tAD1
tAD1
tAD1
A11*
Write A command
tCSD1
tCSD1
CSn
tRWD tRWD tRWD
RD/WR
tRASD tRASD
RAS
tCASD tCASD
CAS
tDQMD tDQMD
DQMxx
tWDD2 tWDH2
D15 to D0
tBSD tBSD
BS
(High)
CKE
Note: * An address pin connected to pin A10 of SDRAM
Figure 21.21 Synchronous DRAM Single Write Bus Cycle (Auto-Precharge, WTRCD = 2 Cycles, TRWL = 1 Cycle)
Rev. 6.00 Jun. 12, 2007 Page 564 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Tr
Tc1
Tc2
Tc3
Tc4
Trwl
CKIO
tAD1 tAD1 Column address tAD1 Column address tAD1 Column address tAD1 Column address tAD1
A25 to A0
Row address
tAD1
tAD1 Write command
tAD1
tAD1
A11*
Write A command
tCSD1
tCSD1
CSn
tRWD tRWD tRWD
RD/WR
tRASD tRASD
RAS
tCASD tCASD
CAS
tDQMD tDQMD
DQMxx
tWDD2 tWDH2 tWDD2 tWDH2
D15 to D0
tBSD tBSD
BS
(High)
CKE
Note: * An address pin connected to pin A10 of SDRAM
Figure 21.22 Synchronous DRAM Burst Write Bus Cycle (Single Write x 4) (Auto-Precharge, WTRCD = 0 Cycle, TRWL = 1 Cycle)
Rev. 6.00 Jun. 12, 2007 Page 565 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Tr
Trw
Tc1
Tc2
Tc3
Tc4
Trwl
CKIO
tAD1 tAD1 Row address Column address tAD1 Column address tAD1 Column address tAD1 Column address tAD1
A25 to A0
tAD1
tAD1 Write command
tAD1
tAD1
A11*
Write A command
tCSD1
tCSD1
CSn
tRWD tRWD tRWD
RD/WR
tRASD tRASD
RAS
tCASD tCASD
CAS
tDQMD tDQMD
DQMxx
tWDD2 tWDH2 tWDD2 tWDH2
D15 to D0
tBSD tBSD
BS
(High)
CKE
Note: * An address pin connected to pin A10 of SDRAM
Figure 21.23 Synchronous DRAM Burst Write Bus Cycle (Single Write x 4) (Auto-Precharge, WTRCD = 1 Cycle, TRWL = 1 Cycle)
Rev. 6.00 Jun. 12, 2007 Page 566 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Td1 Tr Tc1 Tc2 Tc3
Td2 Tc4
Td3
Td4 Tde
CKIO
tAD1 tAD1 Column address tAD1 Column address tAD1 Column address tAD1 Column address tAD1
A25 to A0
Row address
tAD1
tAD1 Read command
tAD1
A11*
tCSD1
tCSD1
CSn
tRWD tRWD
RD/WR
tRASD tRASD
RAS
tCASD tCASD
CAS
tDQMD tDQMD
DQMxx
tRDS2 tRDH2 tRDS2 tRDH2
D15 to D0
tBSD tBSD
BS
(High)
CKE
Note: * An address pin connected to pin A10 of SDRAM
Figure 21.24 Synchronous DRAM Burst Read Bus Cycle (Single Read x 4) (Bank Active Mode: ACT + READ Commands, CAS Latency = 2, WTRCD = 0 Cycle)
Rev. 6.00 Jun. 12, 2007 Page 567 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Td1 Tc1 Tc2 Tc3
Td2 Tc4
Td3
Td4 Tde
CKIO
tAD1 tAD1 Column address tAD1 Column address tAD1 Column address tAD1
A25 to A0
Column address
tAD1
tAD1 Read command
A11*
tCSD1
tCSD1
CSn
tRWD tRWD
RD/WR
tRASD
RAS
tCASD tCASD
CAS
tDQMD tDQMD
DQMxx
tRDS2 tRDH2 tRDS2 tRDH2
D15 to D0
tBSD tBSD
BS
(High)
CKE
Note: * An address pin connected to pin A10 of SDRAM
Figure 21.25 Synchronous DRAM Burst Read Bus Cycle (Single Read x 4) (Bank Active Mode: READ Command, Same Row Address, CAS Latency = 2, WTRCD = 0 Cycle)
Rev. 6.00 Jun. 12, 2007 Page 568 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Td1 Tp Tpw Tr Tc1 Tc2 Tc3
Td2 Tc4
Td3
Td4 Tde
CKIO
tAD1 tAD1 Row address Column address tAD1 Column address tAD1 Column address tAD1 Column address tAD1
A25 to A0
tAD1
tAD1
tAD1 Read command
tAD1
A11*
tCSD1
tCSD1
CSn
tRWD tRWD tRWD
RD/WR
tRASD tRASD tRASD tRASD
RAS
tCASD tCASD
CAS
tDQMD tDQMD
DQMxx
tRDS2 tRDH2 tRDS2 tRDH2
D15 to D0
tBSD tBSD
BS
(High)
CKE
Note: * An address pin connected to pin A10 of SDRAM
Figure 21.26 Synchronous DRAM Burst Read Bus Cycle (Single Read x 4) (Bank Active Mode: PRE + ACT + READ Commands, Different Row Addresses, CAS Latency = 2, WTRCD = 0 Cycle)
Rev. 6.00 Jun. 12, 2007 Page 569 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Tr
Tc1
Tc2
Tc3
Tc4
CKIO
tAD1 tAD1 Row address Column address tAD1 Column address tAD1 Column address tAD1 Column address tAD1
A25 to A0
tAD1
tAD1 Write command
tAD1
A11*
tCSD1
tCSD1
CSn
tRWD tRWD tRWD
RD/WR
tRASD tRASD
RAS
tCASD tCASD
CAS
tDQMD tDQMD
DQMxx
tWDD2 tWDH2 tWDD2 tWDH2
D15 to D0
tBSD tBSD
BS
(High)
CKE
Note: * An address pin connected to pin A10 of SDRAM
Figure 21.27 Synchronous DRAM Burst Write Bus Cycle (Single Write x 4) (Bank Active Mode: ACT + WRITE Commands, WTRCD = 0 Cycle, TRWL = 0 Cycle)
Rev. 6.00 Jun. 12, 2007 Page 570 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Tnop
Tc1
Tc2
Tc3
Tc4
CKIO
tAD1 tAD1 Column address Column address tAD1 Column address tAD1 Column address tAD1
A25 to A0
tAD1 tAD1
tAD1 Write command
A11*
tCSD1
tCSD1
CSn
tRWD tRWD tRWD
RD/WR
RAS
tCASD tCASD
CAS
tDQMD tDQMD
DQMxx
tWDD2 tWDH2 tWDD2 tWDH2
D15 to D0
tBSD tBSD
BS
(High)
CKE
Note: * An address pin connected to pin A10 of SDRAM
Figure 21.28 Synchronous DRAM Burst Write Bus Cycle (Single Write x 4) (Bank Active Mode: WRITE Command, Same Row Address, WTRCD = 0 Cycle, TRWL = 0 Cycle)
Rev. 6.00 Jun. 12, 2007 Page 571 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Tp
Tap
Tr
Tc1
Tc2
Tc3
Tc4
CKIO
tAD1 tAD1 Row address Column address tAD1 Column address tAD1 Column address tAD1 Column address tAD1
A25 to A0
tAD1
tAD1
tAD1 Write command
tAD1
A11*
tCSD1
tCSD1
CSn
tRWD tRWD tRWD tRWD
RD/WR
tRASD tRASD tRASD tRASD
RAS
tCASD tCASD
CAS
tDQMD tDQMD
DQMxx
tWDD2 tWDH2 tWDD2 tWDH2
D15 to D0
tBSD tBSD
BS
(High)
CKE
Note: * An address pin connected to pin A10 of SDRAM
Figure 21.29 Synchronous DRAM Burst Write Bus Cycle (Single Write x 4) (Bank Active Mode: PRE + ACT + WRITE Commands, Different Row Addresses, WTRCD = 0 Cycle, TRWL = 0 Cycle)
Rev. 6.00 Jun. 12, 2007 Page 572 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Tp
Tap
Trr
Trc
Trc
Trc
CKIO
tAD1 tAD1
A25 to A0
tAD1 tAD1
A11*
tCSD1
tCSD1
tCSD1
tCSD1
CSn
tRWD tRWD tRWD
RD/WR
tRASD tRASD tRASD tRASD
RAS
tCASD tCASD
CAS
DQMxx
(Hi-Z)
D15 to D0
BS
(High)
CKE
Note: * An address pin connected to pin A10 of SDRAM
Figure 21.30 Synchronous DRAM Auto-Refreshing Timing (WTRP = 1 Cycle, WTRC = 3 Cycles)
Rev. 6.00 Jun. 12, 2007 Page 573 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Tp
Tap
Trr
Trc
Trc
Trc
Trc
Trc
CKIO
tAD1 tAD1
A25 to A0
tAD1
tAD1
A11*
tCSD1
tCSD1
tCSD1
tCSD1
CSn
tRWD tRWD tRWD
RD/WR
tRASD tRASD tRASD tRASD
RAS
tCASD tCASD
CAS
DQMxx
(Hi-Z)
D15 to D0
BS
tCKED tCKED
CKE
Note: * An address pin connected to pin A10 of SDRAM
Figure 21.31 Synchronous DRAM Self-Refreshing Timing (WTRP = 1 Cycle)
Rev. 6.00 Jun. 12, 2007 Page 574 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Tp
Tap
Trr
Trc
Trc
Trr
Trc
Trc
Tmw
Tde
CKIO
PALL tAD1
REF
REF
MRS tAD1 tAD1
A25 to A0
tAD1 tAD1
A11*
tCSD1
tCSD1
tCSD1
tCSD1
tCSD1
tCSD1
tCSD1
tCSD1
CSn
tRWD tRWD tRWD tRWD tRWD
RD/WR
tRASD tRASD tRASD tRASD tRASD tRASD tRASD tRASD
RAS
tCASD tCASD tCASD tCASD tCASD tCASD
CAS
DQMxx
(Hi-Z)
D15 to D0
BS
CKE
Note: * An address pin connected to pin A10 of SDRAM
Figure 21.32 Synchronous DRAM Mode Register Write Timing (WTRP = 1 Cycle)
Rev. 6.00 Jun. 12, 2007 Page 575 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
21.4.6
PCMCIA Timing
Tpcm1 CKIO tAD1 A25 to A0 tCSD1 CExx tRWD RD/WR tRSD RD tRDH1 tRSD tRWD tCSD1 tAD1 Tpcm1w Tpcm1w Tpcm1w Tpcm2
Read D15 to D0 tWED WE Write D15 to D0 tBSD BS tBSD
tRDS1
tWED
tWDH5 tWDD1 tWDH1
Figure 21.33 PCMCIA Memory Card Interface Bus Timing
Rev. 6.00 Jun. 12, 2007 Page 576 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Tpcm0
Tpcm0w
Tpcm1
Tpcm1w
Tpcm1w
Tpcm1w
Tpcm1w
Tpcm2
Tpcm2w
CKIO
tAD1 tAD1
A25 to A0
tCSD1 tCSD1
CExx
tRWD tRWD
RD/WR
tRSD tRSD
RD Read D15 to D0
tWED tWED
tRDH1 tRDS1
WE Write D15 to D0
tBSD tBSD tWTH tWTS tWTH tWTS tWDD1
tWDH5 tWDH1
BS WAIT
Figure 21.34 PCMCIA Memory Card Interface Bus Timing (TED = 2.5 Cycles, TEH = 1.5 Cycles, One Software Wait Cycle, One External Wait Cycle)
Rev. 6.00 Jun. 12, 2007 Page 577 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Tpci1 CKIO tAD1 A25 to A0 tCSD1 CExx tRWD RD/WR tICRSD ICIORD Read D15 to D0 tICWSD ICIOWR Write D15 to D0 tBSD BS tWDD1
Tpci1w
Tpci1w
Tpci1w
Tpci2
tAD1
tCSD1
tRWD
tICRSD tRDH1 tRDS1
tICWSD tWDH5 tWDH1
tBSD
Figure 21.35 PCMCIA I/O Card Interface Bus Timing
Rev. 6.00 Jun. 12, 2007 Page 578 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Tpci0
Tpci0w
Tpci1
Tpci1w
Tpci1w
Tpci1w
Tpci1w
Tpci2
Tpci2w
CKIO tAD1 A25 to A0 tCSD1 CExx tRWD RD/WR tICRSD ICIORD Read D15 to D0 tICWSD ICIOWR Write D15 to D0 tBSD BS tBSD tWTH tWTS
WAIT
tAD1
tCSD1
tRWD
tICRSD tRDH1 tRDS1
tICWSD tWDH5
tWDD1
tWDH1
tWTH tWTS
tIO16S
IOIS16
tIO16H
Figure 21.36 PCMCIA I/O Card Interface Bus Timing (TED = 2.5 Cycles, TEH = 1.5 Cycles, One Software Wait Cycle, One External Wait Cycle)
Rev. 6.00 Jun. 12, 2007 Page 579 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
21.4.7
SCIF Timing
Table 21.10 SCIF Timing Conditions: VCCQ = 3.0 V to 3.6 V, VCC = 1.4 V to 1.6 V, Ta = -20C to +75C (regular specifications), Ta = -40C to +85C (wide-range specifications)
Item Input clock cycle Clocked synchronous Asynchronous Input clock rising time Input clock falling time Input clock pulse width Transmit data delay time Receive data setup time (clocked synchronous) Receive data hold time (clocked synchronous) RTS delay time CTS setup time (clocked synchronous) CTS hold time (clocked synchronous) Note: * tSCKR tSCKF tSCKW tTXD tRXS tRXH tRTSD tCTSS tCTSH Symbol Min. Max. tScyc 12 4 0.4 3 3 100 100 0.8 0.8 0.6 100 Unit tpcyc tpcyc tpcyc tpcyc tScyc Figure 21.38 tpcyc tpcyc ns ns ns Figure 21.37 Reference Figures Figures 21.37 and 21.38
3 x tpcyc* + 50 ns
tpcyc indicates the period of the peripheral module clock (P).
tSCKW SCK
tSCKR
tSCKF
tScyc
Figure 21.37 SCK Input Clock Timing
Rev. 6.00 Jun. 12, 2007 Page 580 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
tScyc SCK tTXD TxD (data transmission) tRXS RxD (data reception) tRTSD RTS tCTSS CTS tCTSH tRXH
Figure 21.38 SCI Input/Output Timing in Clocked Synchronous Mode 21.4.8 Port Timing
Table 21.11 Port Timing Conditions: VCCQ = 3.0 V to 3.6 V, VCC = 1.4 V to 1.6 V, Ta = -20C to +75C (regular specifications), Ta = -40C to +85C (wide-range specifications)
Item Output data delay time Input data setup time Input data hold time Symbol tPORTD tPORTS tPORTH Min. 16 10 Max. 20 Unit ns ns ns Reference Figures Figure 21.39
CKIO tPORTS tPORTH
Ports A to E (read)
tPORTD
Ports A to E (write)
Figure 21.39 I/O Port Timing
Rev. 6.00 Jun. 12, 2007 Page 581 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
21.4.9
HIF Timing
Table 21.12 HIF Timing Conditions: VCCQ = 3.0 V to 3.6 V, VCC = 1.4 V to 1.6 V, Ta = -20C to +75C (regular specifications), Ta = -40C to +85C (wide-range specifications)
Item Read bus cycle time Write bus cycle time Address setup time (HIFSCR.DMD = 0) Address setup time (HIFSCR.DMD = 1) Address hold time (HIFSCR.DMD = 0) Address hold time (HIFSCR.DMD = 1) Read low width (read) Write low width (write) Read/write high width Read data delay time Read data hold time Write data setup time Write data hold time HIFINT output delay time HIFRDY output delay time HIFDREQ output delay time HIF pin enable delay time HIF pin disable delay time Symbol Min. tHIFCYCR tHIFCYCW tHIFAS tHIFAS tHIFAH tHIFAH tHIFWRL tHIFWWL tHIFWRWH tHIFRDD tHIFRDH tHIFWDS tHIFWDH tHIFITD tHIFRYD tHIFDQD tHIFEBD tHIFDBD 4 4 10 0 10 0 2.5 2.5 2.0 0 tpcyc + 10 10 Max. 2 x tpcyc + 16 20 10 20 20 20 Unit tpcyc tpcyc ns ns ns ns tpcyc tpcyc tpcyc ns ns ns ns ns tpcyc ns ns ns Figure 21.41 Figure 21.42 Figure 21.41 Figure 21.42 Reference Figures Figure 21.40
Notes: 1. tpcyc indicates the period of the peripheral module clock (P). 2. tHIFAS is given from the start of the time over which both the HIFCS and HIFRD (or HIFWR) signals are low levels. 3. tHIFAH is given from the end of the time over which both the HIFCS and HIFRD (or HIFWR) signals are low levels. 4. tHIFWRL is given as the time over which both the HIFCS and HIFRD signals are low levels. 5. tHIFWWL is given as the time over which both the HIFCS and HIFWR signals are low levels. 6. When reading the register specified by bits REG5 to REG0 after writing to the HIF index register (HIFIDX), tHIFWRWH (min.) = 2 x tpcyc + 5 ns.
Rev. 6.00 Jun. 12, 2007 Page 582 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
tHIFCYCR
HIFRS
tHIFCYCW
tHIFWRL
HIFCS
tHIFWWL
tHIFAS
HIFRD HIFWR
tHIFAH tHIFWRWH tHIFRDD tHIFRDH
Read data
tHIFAS
tHIFAH
tHIFWDS tHIFWDH
Write data
HIFD15 to HIFD00
Figure 21.40 HIF Access Timing
CKIO tHIFITD HIFINT tHIFDQD HIFDREQ
Figure 21.41 HIFINT and HIFDREQ Timing
Rev. 6.00 Jun. 12, 2007 Page 583 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
HIFEBL HIFINT HIFDREQ HIFRDY HIFD15 to HIFD0 RES tHIFRYD HIFRDY tHIFRYD tHIFDBD tHIFEBD
Figure 21.42 HIFRDY and HIF Pin Enable/Disable Timing
Rev. 6.00 Jun. 12, 2007 Page 584 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
21.4.10 EtherC Timing Table 21.13 EtherC Timing Conditions: VCCQ = 3.0 V to 3.6 V, VCC = 1.4 V to 1.6 V, Ta = -20C to +75C (regular specifications), Ta = -40C to +85C (wide-range specifications)
Item TX-CLK cycle time TX-EN output delay time MII_TXD[3:0] output delay time CRS setup time CRS hold time COL setup time COL hold time RX-CLK cycle time RX-DV setup time RX-DV hold time MII_RXD[3:0] setup time MII_RXD[3:0] hold time RX-ER setup time RX-ER hold time MDIO setup time MDIO hold time MDIO output data hold time WOL output delay time EXOUT output delay time Symbol tTcyc tTENd tMTDd tCRSs tCRSh tCOLs tCOLh tRcyc tRDVs tRDVh tMRDs tMRDh tRERs tRERh tMDIOs tMDIOh tMDIOdh tWOLd tEXOUTd Min. 40 1 1 10 10 10 10 40 10 10 10 10 10 10 10 10 5 1 1 Max. 20 20 18 20 20 Unit ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns Figure 21.48 Figure 21.49 Figure 21.50 Figure 21.47 Figure 21.46 Figure 21.45 Figure 21.44 Reference Figures Figure 21.43
Rev. 6.00 Jun. 12, 2007 Page 585 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
TX-CLK tTENd TX-EN tMTDd MII_TXD[3:0] Preamble SFD DATA CRC
TX-ER tCRSs CRS tCRSh
COL
Figure 21.43 MII Transmission Timing (Normal Operation)
TX-CLK
TX-EN
MII_TXD[3:0]
Preamble
JAM
TX-ER
CRS tCOLs COL tCOLh
Figure 21.44 MII Transmission Timing (Collision Occurred)
Rev. 6.00 Jun. 12, 2007 Page 586 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
RX-CLK tRDVs RX-DV tMRDs MII_RXD[3:0] Preamble SFD DATA CRC tMRDh tRDVh
RX-ER
Figure 21.45 MII Reception Timing (Normal Operation)
RX-CLK
RX-DV
MII_RXD[3:0]
Preamble
SFD
DATA tRERs tRERh
xxxx
RX-ER
Figure 21.46 MII Reception Timing (Error Occurred)
MDC tMDIOs MDIO tMDIOh
Figure 21.47 MDIO Input Timing
MDC tMDIOdh MDIO
Figure 21.48 MDIO Output Timing
Rev. 6.00 Jun. 12, 2007 Page 587 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
RX-CLK tWOLd WOL
Figure 21.49 WOL Output Timing
CKIO tEXOUTd EXOUT
Figure 21.50 EXOUT Output Timing 21.4.11 H-UDI Related Pin Timing Table 21.14 H-UDI Related Pin Timing Conditions: VCCQ = 3.0 V to 3.6 V, VCC = 1.4 V to 1.6 V, Ta = -20C to +75C (regular specifications), Ta = -40C to +85C (wide-range specifications)
Item TCK cycle time TCK high pulse width TCK low pulse width TCK rising/falling time TRST setup time TRST hold time TDI setup time TDI hold time TMS setup time TMS hold time TDO delay time Note: * Symbol tTCKcyc tTCKH tTCKL tTCKrf tTRSTS tTRSTH tTDIS tTDIH tTMSS tTMSH tTDOD Min. 50 19 19 10 50 10 10 10 10 Max. 4 19 Unit ns ns ns ns tbcyc* tbcyc* ns ns ns ns ns Figure 21.53 Figure 21.52 Reference Figures Figure 21.51
tbcyc indicates the period of the external bus clock (B).
Rev. 6.00 Jun. 12, 2007 Page 588 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
tTCKcyc tTCKH tTCKL VIH VIH 1/2 VccQ VIH 1/2 VccQ VIL tTCKrf VIL tTCKrf
Figure 21.51 TCK Input Timing
RES tTRSTS TRST tTRSTH
Figure 21.52 TCK Input Timing in Reset Hold State
tTCKcyc
TCK tTDIS TDI tTMSS TMS tTDOD TDO tTMSH tTDIH
Figure 21.53 H-UDI Data Transmission Timing
Rev. 6.00 Jun. 12, 2007 Page 589 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
21.4.12 AC Characteristic Test Conditions * I/O signal reference level: VCCQ/2 (VCCQ = 3.0 V to 3.6 V, VCC = 1.4 V to 1.6 V) * Input pulse level: VSS to VCC (RES, NMI, IRQ7 to IRQ0, MD5, MD3 to MD0, ASEMD, TESTMD, HIFMD, TRST, and EXTAL), VSS to 3.0 V (other pins) * Input rising and falling times: 1 ns
IOL
LSI output pin CL
DUT output VREF
IOH
Notes: 1. CL is the total value that includes the capacitance of measurement instruments, etc., and is set for all pins as 30 pF. 2. IOL and IOH are shown in table 21.5.
Figure 21.54 Output Load Circuit
Rev. 6.00 Jun. 12, 2007 Page 590 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
21.4.13 Delay Time Variation Due to Load Capacitance (Reference Values) A graph (reference data) of the variation in delay time when a load capacitance greater than that stipulated (30 pF) is connected to this LSI's pins is shown below. The graph shown in figure 21.55 should be taken into consideration in the design process if the stipulated capacitance is exceeded in connecting an external device. If the connected load capacitance exceeds the range shown in figure 21.55, the graph will not be a straight line.
+3
Delay time (ns)
+2
+1
+0 +0 +10 +20 +30 +40 +50 Load capacitance (pF)
Figure 21.55 Load Capacitance versus Delay Time
Rev. 6.00 Jun. 12, 2007 Page 591 of 610 REJ09B0131-0600
Section 21 Electrical Characteristics
Rev. 6.00 Jun. 12, 2007 Page 592 of 610 REJ09B0131-0600
Appendix
Appendix
A. Port States in Each Pin State
Port States in Each Pin State
Reset State Classification Clock Power-On (HIFMD = Low) I O* O* O I I
1
Table A.1
Power-Down Mode Software Standby I H-UDI Module Standby I
1
Abbr. EXTAL XTAL CKIO CK_PHY
Power-On (HIFMD = High) I O* O* O I I
1
Sleep I
O*
1
O*
5
O* O* O I I
1
1
1
ZO* H I I
O* O I I
1
1
System control
RES
Operating MD5, MD3 to MD0 mode control Interrupt NMI IRQ4 to IRQ0 Address bus Data bus Bus control A25 to A16 A15 to A0 D15 to D0 WAIT IOIS16 CKE CAS, RAS WE0/DQMLL
I O Z H
I O Z H H H
I I ZHL* ZHL* Z Z Z ZO* ZO* ZH*
2 4
I I O O IO I I O O O O O O O
I I O O IO I I O O O O O O O
4
2
4
WE1/DQMLU/ H WE ICIORD ICIOWR RD H
ZH*
4
ZH* ZH* ZH*
4
4
4
Rev. 6.00 Jun. 12, 2007 Page 593 of 610 REJ09B0131-0600
Appendix
Reset State Classification Bus control Power-On (HIFMD = Low) H H Power-On (HIFMD = High) H H Software Standby ZH* ZH* ZH*
4
Power-Down Mode H-UDI Module Standby O O O O O O I O I I I O O I I I IO O I O O O I O I O
Abbr. RD/WR CE2B/CE2A CS6B/CE1B, CS5B/CE1A CS4, CS3 CS0 BS
Sleep O O O O O O I O I I I O O I I I IO O I O O O I O I O
4
4
ZH* ZH*
4
4
ZH* I O I I I O O I I I IO O Z Z Z Z Z Z Z Z
4
Ethernet controller
MII_RXD3 to MII_RXD0 MII_TXD3 to MII_TXD0 RX_DV RX_ER RX_CLK TX_ER TX_EN TX_CLK COL CRS MDIO MDC LNKSTA EXOUT WOL
SCIF
TXD2 to TXD0 RXD2 to RXD0 SCK2, SCK1 SCK0 RTS1, RTS0
Rev. 6.00 Jun. 12, 2007 Page 594 of 610 REJ09B0131-0600
Appendix
Reset State Classification SCIF Host interface Power-On (HIFMD = Low) I I I I I Z I Power-On (HIFMD = High) Z O Z I Z Z Z Z Z Z I I I I Z I Z Z Z Z I O Software Standby Z Z O Z I Z Z Z Z Z Z I I I I ZO* I Z Z Z Z Z Z I O
6
Power-Down Mode H-UDI Module Standby I I
3
Abbr. CTS1, CTS0 HIFEBL HIFRDY HIFDREQ HIFMD HIFINT HIFRD HIFWR HIFRS HIFCS HIFD15 to HIFD0
Sleep I I O* O* I*
3
O* O* I*
3
3
3
3
O* I*
3
3
O* I* I* I* I*
3 3
3
I* I* I*
3
3
3
3
3
3
IO* I I I I
IO* I I I I
6
3
User TRST debugging TCK interface TMS (H-UDI) TDI TDO ASEMD I/O port
ZO* I P P P P P P I O
Z I I/O I/O I/O I/O I/O I/O I O
PA25 to PA16 Z PB13 to PB00 Z PC20 to PC00 Z PD07 to PD00 Z PE24 to PE04, Z PE02 to PE00 PE03 I O
Test mode TESTMD TESTOUT
[Legend] : This pin function is not selected as an initial state. I: Input O: Output
Rev. 6.00 Jun. 12, 2007 Page 595 of 610 REJ09B0131-0600
Appendix
IO: H: L: Z: P: Notes:
Input/output High level output Low level output High-impedance Input or output depending on the register setting 1. Depends on the clock mode (setting of pins MD2 to MD0). 2. Depends on the HIZCNT bit in CMNCR. 3. High-impedance when HIFEBL = low 4. Depends on the HIZMEM bit in CMNCR. 5. Depends on the HIZCNT bit in CMNCR or the CKOEN bit in FRQCR. 6. This pin becomes output state only when reading data from the H-UDI and retains highimpedance state when the pin is not output state.
Rev. 6.00 Jun. 12, 2007 Page 596 of 610 REJ09B0131-0600
Appendix
B.
Product Code Lineup
* SH7618
Product Code D17618RBG100V D17618RBGN100V Catalogue Code Operating Temperature Solder Ball HD6417618R BG100V HD6417618R BGN100V -20 to 75C -20 to 75C -40 to 85C -20 to 75C -20 to 75C -40 to 85C Pb-free Pb-free Pb-free Non-Pb-free Non-Pb-free Non-Pb-free Package Code PLBG0176GAA
D17618RBGW100V HD6417618R BGW100V D17618RBG100 D17618RBGN100 D17618RBGW100 HD6417618R BG100 HD6417618R BGN100 HD6417618R BGW100
* SH7618A
Product Code D17618ABG100V D17618ABGN100V Catalogue Code Operating Temperature Solder Ball HD6417618A BG100V HD6417618A BGN100V -20 to 75C -20 to 75C -40 to 85C -20 to 75C -20 to 75C -40 to 85C Pb-free Pb-free Pb-free Non-Pb-free Non-Pb-free Non-Pb-free Package Code PLBG0176GAA
D17618ABGW100V HD6417618A BGW100V D17618ABG100 D17618ABGN100 D17618ABGW100 HD6417618A BG100 HD6417618A BGN100 HD6417618A BGW100
Rev. 6.00 Jun. 12, 2007 Page 597 of 610 REJ09B0131-0600
Appendix
C.
Package Dimensions
RENESAS Code PLBG0176GA-A D wSA wSB Previous Code BP-176/BP-176V MASS[Typ.] 0.45g
JEITA Package Code P-LFBGA176-13x13-0.80
x4
v y1 S
S
E
y S e A ZD
Reference Dimension in Millimeters Symbol Min Nom Max
A1
A
e
R P N M L K J H G F E D
D E v B w A A1 e b x 0.45 0.35
13.0 13.0 0.15 0.20 1.40 0.40 0.80 0.50 0.45 0.55 0.08 0.10 0.2
ZE
C B A
y y1 SD
1
2
3
4
5
6
7
8
9 10 11 12 13 14 15
SE ZD ZE 0.90 0.90
b
x M S A B
Figure C.1 Package Dimensions (BP-176)
Rev. 6.00 Jun. 12, 2007 Page 598 of 610 REJ09B0131-0600
Main Revisions and Additions in this Edition
Item 1. Auto-refreshing Page Revision (See Manual for Details) 175 Amended : set so as to satisfy the SDRAM refreshing cycle time (tRC). A Tpw cycle is inserted between the Tp cycle and Trr cycle when the setting of bits WTRP1 and WTRP0 in CSnWCR is longer than or equal to one cycle. 14.3.7 Serial Status Register (SCFSR) 341 Amended
Mode 5 Description Transmit FIFO Data Empty Indicates that data has been transferred from the transmit FIFO data register (SCFTDR) to the transmit shift register (SCTSR), the quantity of data in SCFTDR has become less than the transmission trigger number specified by the TTRG1 and TTRG0 bits in the FIFO control register (SCFCR), and writing of transmit data to SCFTDR is enabled. 0: The quantity of transmit data written to SCFTDR is greater than the specified transmission trigger number [Clearing conditions] * TDFE is cleared to 0 when data exceeding the specified transmission trigger number is written to SCFTDR after 1 is read from TDFE and then 0 is written TDFE is cleared to 0 when DMAC write data exceeding the specified transmission trigger number to SCFTDR
*
1: The quantity of transmit data in SCFTDR is equal to or less than the specified transmission trigger number* [Setting conditions] * * TDFE is set to 1 by a power-on reset TDFE is set to 1 when the quantity of transmit data in SCFTDR has become equal to or less than the specified transmission trigger number as a result of transmission Since SCFTDR is a 16-byte FIFO register, the
Note: *
maximum quantity of data that can be written when TDFE is 1 is "16 minus the specified transmission trigger number". If an attempt is made to write additional data, the data is ignored. The quantity of data in SCFTDR is indicated by the upper 8 bits of SCFDR.
Rev. 6.00 Jun. 12, 2007 Page 599 of 610 REJ09B0131-0600
Item 14.4.3 Synchronous Operation Clock
Page Revision (See Manual for Details) 376 Amended : When only receiving, the clock signal outputs while the RE bit of SCSCR is 1 and the number of data in receive FIFO is less than the receive FIFO data trigger number. In this case, 8 x (16 + 1) = 136 pulses of synchronous clock are output. To perform reception of n characters of data, select an external clock as the clock source. If an internal clock should be used, set RE = 1 and TE = 1 and receive n characters of data simultaneously with the transmission of n characters of dummy data.
Figure 14.13 Sample Flowchart for Transmitting Serial Data
378
Amended
Start of transmission [1] SCIF status check and transmit data write: No Read SCFSR and check that the TDFE flag is set to 1, then write transmit data to SCFTDR. Read the TDFE and TEND flags while they are 1, then clear them to 0. [2] Serial transmission continuation procedeure: To continue serial transmission, read 1 from the TDFE flag to confirm that writing is possible, them write data to
Read TDFE flag in SCFSR
TDFE = 1? Yes Write transmit data to SCFTDR Read TDFE and TEND flags in SCFSR while they are 1, then clear them to 0
[1]
Figure 14.18 Sample Flowchart for Transmitting/Receiving Serial Data
382
Amended
Initialization [1] SCIF status check and transmit data write: Read SCFSR and check that the TDFE flag is set to 1, then write transmit data to SCFTDR. Read the TDFE and TEND flags while they are 1, then clear them to 0. The transition of the TDFE flag from 0 to 1 can also be identified by a TXI interrupt. [2] Receive error handling: TDFE = 1? Yes Write transmit data to SCFTDR Read TDFE and TEND flags in SCFSR while they are 1, then clear them to 0 Read the ORER flag in SCLSR to identify any error, perform the appropriate error handling, then clear the ORER flag to 0. Reception cannot be resumed while the ORER flag is set to 1. [3] SCIF status check and receive data read: Start of transmission and reception
Read TDFE flag in SCFSR
No
[1]
19.6 Usage Notes
470
Added 2. Since the HIFMD pin is not initially set to function as a general port pin, it must be pulled up or down externally to fix its state. 3. When using a multiplexed pin with a function not selected with its initial value (for example, using the PB12/CS3 pin, the initial function of which is PB12, as the CS3 pin), the pin must be pulled up or down externally at least after a reset until its pin function is selected by software to fix its state.
Rev. 6.00 Jun. 12, 2007 Page 600 of 610 REJ09B0131-0600
Item Figure 21.10 Basic Bus Timing: No Wait Cycle
Page Revision (See Manual for Details) 553 Amended
T1 T2
CKIO
tAD1 tAD1
A25 to A0
tAS tCSD1 tCSD1
CSn
tRWD
tRWD
RD/WR
Figure 21.11 Basic Bus Timing: One Software Wait Cycle
554
Amended
T1 Tw T2
CKIO
tAD1 tAD1
A25 to A0
tAS tCSD1 tCSD1
CSn
tRWD
tRWD
RD/WR
tRSD tAH
Figure 21.12 Basic Bus Timing: One External Wait Cycle
555
Amended
T1 TwX T2
CKIO
tAD1 tAD1
A25 to A0
tAS tCSD1 tCSD1
CSn
tRWD tRWD
RD/WR
tRSD tAH
Figure 21.13 Basic Bus Timing: One Software Wait Cycle, External Wait Enabled (WM Bit = 0), No Idle Cycle
556
Amended
T1 Tw T2 Taw T1 Tw T2 Taw
CKIO
tAD1 tAD1 tAD1 tAD1
A25 to A0
tAS tCSD1 tCSD1 tAS tCSD1 tCSD1
CSn
tRWD tRWD tRWD tRWD
RD/WR
Rev. 6.00 Jun. 12, 2007 Page 601 of 610 REJ09B0131-0600
Item Figure 21.16 Synchronous DRAM Single Read Bus Cycle (Auto-Precharge, CAS Latency = 2, WTRCD = 0 Cycle, WTRP = 0 Cycle) Figure 21.17 Synchronous DRAM Single Read Bus Cycle (Auto-Precharge, CAS Latency = 2, WTRCD = 1 Cycle, WTRP = 1 Cycle) Figure 21.18 Synchronous DRAM Burst Read Bus Cycle (Single Read x 4) (Auto-Precharge, CAS Latency = 2, WTRCD = 0 Cycle, WTRP = 1 Cycle) Figure 21.19 Synchronous DRAM Burst Read Bus Cycle (Single Read x 4) (Auto-Precharge, CAS Latency = 2, WTRCD = 1 Cycle, WTRP = 0 Cycle) Figure 21.20 Synchronous DRAM Single Write Bus Cycle (Auto-Precharge, TRWL = 1 Cycle)
Page Revision (See Manual for Details) 559 Amended
Tr Tc1 Tcw Td1 Tde
CKIO tAD1 A25 to A0 tAD1
Row address Column address
tAD1
560
Amended
Tr Trw Tc1 Tcw Td1 Tde Tap
CKIO
tAD1 tAD1
Row address Column address
tAD1
A25 to A0
561
Amended
Td1 Tr TC1 TC2 Tc4 Td2 Tc4 Td3 Td4 Tde
CKIO
tAD1
tAD1 Column address
tAD1 Column address
tAD1 Column address
tAD1 Column address
tAD1
A25 to A0
Row address
562
Amended
Td1 Tr Trw Tc1 Tc2 Tc3 Td2 Tc4 Td3 Td4 Tde
CKIO
tAD1
tAD1 Row address Column address
tAD1 Column address
tAD1 Column address
tAD1 Column address
tAD1
A25 to A0
563
Amended
Tr Tc1 Trwl
CKIO
tAD1
tAD1
tAD1
A25 to A0
Row address Column address
Figure 21.21 Synchronous DRAM Single Write Bus Cycle (Auto-Precharge, WTRCD = 2 Cycles, TRWL = 1 Cycle)
564
Amended
Tr Trw Trw Tc1 Trwl
CKIO
tAD1 tAD1
Row address
tAD1
A25 to A0
Column address
Figure 21.22 Synchronous DRAM Burst Write Bus Cycle (Single Write x 4) (Auto-Precharge, WTRCD = 0 Cycle, TRWL = 1 Cycle)
565
Amended
Tr Tc1 Tc2 Tc3 Tc4 Trwl
CKIO
tAD1
tAD1 Row address Column address
tAD1 Column address
tAD1 Column address
tAD1 Column address
tAD1
A25 to A0
Rev. 6.00 Jun. 12, 2007 Page 602 of 610 REJ09B0131-0600
Item Figure 21.23 Synchronous DRAM Burst Write Bus Cycle (Single Write x 4) (Auto-Precharge, WTRCD = 1 Cycle, TRWL = 1 Cycle)
Page Revision (See Manual for Details) 566 Amended
Tr Trw Tc1 Tc2 Tc3 Tc4 Trwl
CKIO
tAD1
tAD1 Row address Column address
tAD1 Column address
tAD1 Column address
tAD1 Column address
tAD1
A25 to A0
Figure 21.24 Synchronous 567 DRAM Burst Read Bus Cycle (Single Read x 4) (Bank Active Mode: ACT + READ Commands, CAS Latency = 2, WTRCD = 0 Cycle) Figure 21.25 Synchronous DRAM Burst Read Bus Cycle (Single Read x 4) (Bank Active Mode: READ Command, Same Row Address, CAS Latency = 2, WTRCD = 0 Cycle) 568
Amended
Td1 Tr Tc1 Tc2 Tc3 Td2 Tc4 Td3 Td4 Tde
CKIO
tAD1
tAD1 Column address
tAD1 Column address
tAD1 Column address
tAD1 Column address
tAD1
A25 to A0
Row address
Amended
Td1 Tc1 Tc2 Tc3 Td2 Tc4 Td3 Td4 Tde
CKIO
tAD1
tAD1 Column address
tAD1 Column address
tAD1 Column address
tAD1
A25 to A0
Column address
Figure 21.26 Synchronous 569 DRAM Burst Read Bus Cycle (Single Read x 4) (Bank Active Mode: PRE + ACT + READ Commands, Different Row Addresses, CAS Latency = 2, WTRCD = 0 Cycle) Figure 21.27 Synchronous 570 DRAM Burst Write Bus Cycle (Single Write x 4) (Bank Active Mode: ACT + WRITE Commands, WTRCD = 0 Cycle, TRWL = 0 Cycle) Figure 21.28 Synchronous DRAM Burst Write Bus Cycle (Single Write x 4) (Bank Active Mode: WRITE Command, Same Row Address, WTRCD = 0 Cycle, TRWL = 0 Cycle) 571
Amended
Td1 Tp Tpw Tr Tc1 Tc2 Tc3 Td2 Tc4 Td3 Td4 Tde
CKIO
tAD1
tAD1 Row address Column address
tAD1 Column address
tAD1 Column address
tAD1 Column address
tAD1
A25 to A0
Amended
Tr Tc1 Tc2 Tc3 Tc4
CKIO
tAD1
tAD1 Row address Column address
tAD1 Column address
tAD1 Column address
tAD1 Column address
tAD1
A25 to A0
Amended
Tnop Tc1 Tc2 Tc3 Tc4
CKIO
tAD1
tAD1 Column address Column address
tAD1 Column address
tAD1 Column address
tAD1
A25 to A0
Rev. 6.00 Jun. 12, 2007 Page 603 of 610 REJ09B0131-0600
Item
Page Revision (See Manual for Details) Amended
Tp Tpw Tr Tc1 Tc2 Tc3 Tc4
Figure 21.29 Synchronous 572 DRAM Burst Write Bus Cycle (Single Write x 4) (Bank Active Mode: PRE + ACT + WRITE Commands, Different Row Addresses, WTRCD = 0 Cycle, TRWL = 0 Cycle) Figure 21.30 Synchronous DRAM Auto-Refreshing Timing (WTRP = 1 Cycle, WTRC = 3 Cycles) 573
CKIO
tAD1
tAD1 Row address Column address
tAD1 Column address
tAD1 Column address
tAD1 Column address
tAD1
A25 to A0
Amended
Tp Tap Trr Trc Trc Trc
CKIO
tAD1 tAD1
A25 to A0
Figure 21.31 Synchronous DRAM Self-Refreshing Timing (WTRP = 1 Cycle)
574
Amended
Tp Tap Trr Trc Trc Trc Trc Trc
CKIO
tAD1 tAD1
A25 to A0
Figure 21.32 Synchronous DRAM Mode Register Write Timing (WTRP = 1 Cycle)
575
Amended
Tp Tap Trr Trc Trc Trr Trc Trc Tmw Tde
CKIO
PALL
REF
REF
MRS
tAD1
tAD1
tAD1
A25 to A0
Rev. 6.00 Jun. 12, 2007 Page 604 of 610 REJ09B0131-0600
Index
A
Access wait control................................. 150 Accessing MII registers .......................... 252 Address array............................................ 58 Address error exception handling ............. 71 Address error sources ............................... 71 Address multiplexing.............................. 154 Addressing modes..................................... 28 Arithmetic operation instructions ............. 41 Asynchronous Mode............................... 362 Auto-refreshing....................................... 174 Coherency of cache and external memory ..................................................... 57 Compare match timer (CMT) ................. 315 Connection to PHY-LSI.......................... 257 Control registers........................................ 21 CPU........................................................... 19
D
Data array.................................................. 59 Data register............................................ 457 Data transfer instructions .......................... 39 Divided areas and cache............................ 51
B
Bank active ............................................. 167 Basic timing............................................ 146 Basic timing for I/O card interface ......... 187 Basic timing for memory card interface . 185 Bit rate .................................................... 346 Boundary scan ........................................ 507 Branch instructions ................................... 44 Burst read................................................ 161 Burst write .............................................. 165 Bus state controller (BSC) ...................... 105 Byte-selection SRAM interface .............. 179
E
Endian/access size and data alignment ... 141 EtherC receiver ....................................... 249 EtherC transmitter................................... 247 Ethernet controller (EtherC).................... 227 Ethernet controller direct memory access controller (E-DMAC) .................. 259 Exception handling ................................... 65 Exception handling operations.................. 66 Exception handling vector table................ 67 Extension of chip select (CSn) assertion period ....................................... 152
C
Cache ........................................................ 49 Cache structure ......................................... 49 Cases when exceptions are accepted......... 76 Changing clock operating mode ............. 203 Changing division ratio........................... 203 Changing frequency................................ 202 Changing multiplication ratio ................. 202 Clock operating modes ........................... 196 Clock Pulse Generator (CPG)................. 193
F
Features of instructions ............................. 25 Flow control............................................ 256
G
General illegal instructions ....................... 75 General registers (Rn) ............................... 21
Rev. 6.00 Jun. 12, 2007 Page 605 of 610 REJ09B0131-0600
H
Host interface (HIF)................................ 391 H-UDI Interrupt...................................... 506 H-UDI Reset........................................... 506
N
NMI interrupt............................................ 95 Normal space interface ........................... 146
O
On-chip peripheral module interrupts ....... 97 Operation by IPG setting ........................ 256
I
I/O ports.................................................. 457 Illegal slot instructions.............................. 75 Immediate data formats ............................ 25 Initial values of registers........................... 23 Instruction formats.................................... 31 Instruction set ........................................... 35 Interrupt controller (INTC)....................... 81 Interrupt exception handling..................... 73 Interrupt exception handling vector table ............................................... 98 Interrupt priority ....................................... 73 Interrupt response time ........................... 102 Interrupt sequence................................... 100 Interrupt sources ....................................... 72 IRQ7 to IRQ0 Interrupts........................... 96
P
PCMCIA interface .................................. 183 Pin assignments........................................... 7 Pin function controller (PFC).................. 423 Pin functions ............................................... 8 Power-down modes................................. 215 Power-on reset .......................................... 69 Power-on sequence ................................. 177
R
Read access............................................... 56 Receive descriptor 0 (RD0) .................... 288 Receive descriptor 1 (RD1) .................... 291 Receive descriptor 2 (RD2) .................... 291 Receiving serial data (asynchronous mode) .............................. 371 Receiving serial data (synchronous mode)................................ 380 Refreshing............................................... 174 Register APR............................. 245, 515, 530, 538 BAMRA...................... 474, 515, 532, 538 BAMRB...................... 476, 515, 531, 538 BARA ......................... 473, 515, 532, 538 BARB ......................... 475, 515, 531, 538 BBRA ......................... 474, 515, 532, 538 BBRB.......................... 477, 515, 532, 538 BDMRB...................... 477, 515, 531, 538 BDRB ......................... 476, 515, 531, 538 BETR .......................... 482, 515, 531, 538 BRCR.......................... 479, 515, 531, 538
L
Logic operation instructions ..................... 43
M
Magic Packet detection........................... 255 Memory data formats................................ 24 Memory-mapped cache ............................ 58 MII frame timing .................................... 250 Module standby mode ............................ 225 Multi-buffer frame transmit/receive processing ............................................... 295 multiplexed pin....................................... 423
Rev. 6.00 Jun. 12, 2007 Page 606 of 610 REJ09B0131-0600
BRDR ......................... 483, 515, 532, 538 BRSR.......................... 482, 515, 532, 538 CCR1 ............................ 52, 515, 532, 538 CCR3 ............................ 53, 510, 516, 533 CDCR ......................... 242, 515, 529, 537 CEFCR ....................... 243, 515, 529, 537 CMCNT .............................................. 318 CMCNT_0 .......................... 512, 521, 535 CMCNT_1 .......................... 513, 521, 535 CMCOR.............................................. 318 CMCOR_0.......................... 512, 521, 535 CMCOR_1.......................... 513, 521, 535 CMCSR .............................................. 317 CMCSR_0 .......................... 512, 520, 535 CMCSR_1 .......................... 513, 521, 535 CMNCR...................... 114, 513, 522, 536 CMSTR....................... 316, 512, 520, 535 CNDCR ...................... 242, 515, 529, 537 CS0BCR ............................. 513, 523, 536 CS0WCR .................... 120, 513, 523, 536 CS3BCR ............................. 513, 523, 536 CS3WCR .....122, 130, 513, 523, 524, 536 CS4BCR ............................. 513, 523, 536 CS4WCR .................... 123, 513, 524, 536 CS5BBCR........................... 513, 523, 536 CS5BWCR...........126, 132, 513, 524, 536 CS6BBCR........................... 513, 523, 536 CS6BWCR...........128, 132, 513, 524, 536 CSnBCR ............................................. 115 CSnWCR ............................................ 120 ECMR......................... 232, 514, 528, 537 ECSIPR....................... 237, 514, 528, 537 ECSR .......................... 235, 514, 528, 537 EDMR......................... 261, 513, 525, 536 EDOCR....................... 279, 514, 527, 537 EDRRR....................... 263, 514, 525, 536 EDTRR ....................... 262, 513, 525, 536 EESIPR....................... 270, 514, 526, 537 EESR .......................... 265, 514, 526, 537 FCFTR........................ 281, 514, 527, 537
FDR............................. 277, 514, 526, 537 FRECR........................ 243, 515, 529, 537 FRQCR ....................... 198, 511, 518, 534 HIFADR...................... 404, 513, 522, 536 HIFBCR ...................... 405, 513, 522, 536 HIFBICR..................... 407, 513, 522, 536 HIFDATA................... 405, 513, 522, 536 HIFDTR ...................... 406, 513, 522, 536 HIFEICR..................... 403, 513, 522, 536 HIFGSR ...................... 398, 513, 521, 535 HIFIDX....................... 395, 513, 521, 535 HIFIICR ...................... 403, 513, 521, 536 HIFMCR ..................... 401, 513, 521, 535 HIFSCR ...................... 398, 513, 521, 535 ICR0.............................. 84, 511, 518, 534 IPGR ........................... 245, 515, 530, 538 IPR ........................................................ 93 IPRA ................................... 511, 518, 534 IPRB.................................... 511, 518, 534 IPRC.................................... 511, 518, 534 IPRD ................................... 511, 518, 534 IPRE.................................... 511, 518, 534 IRQCR .......................... 85, 511, 518, 534 IRQSR........................... 88, 511, 518, 534 LCCR .......................... 242, 515, 529, 537 MAFCR ...................... 244, 515, 530, 538 MAHR ........................ 239, 514, 528, 537 MALR......................... 239, 514, 528, 537 MCLKCR.................... 200, 511, 518, 534 MPR ............................ 246, 515, 530, 538 PACRH1 ..................... 433, 510, 516, 533 PACRH2 ..................... 433, 510, 516, 533 PADRH....................... 457, 510, 516, 533 PAIORH...................... 433, 510, 516, 533 PBCRL1...................... 436, 510, 516, 533 PBCRL2...................... 436, 510, 516, 533 PBDRL........................ 459, 510, 516, 533 PBIORL ...................... 436, 510, 516, 533 PCCRH2 ..................... 440, 510, 517, 533 PCCRL1...................... 440, 510, 517, 533
Rev. 6.00 Jun. 12, 2007 Page 607 of 610 REJ09B0131-0600
PCCRL2 ......................440, 510, 517, 533 PCDRH........................462, 510, 516, 533 PCDRL ........................462, 510, 516, 533 PCIORH ......................440, 510, 516, 533 PCIORL.......................440, 510, 517, 533 PDCRL2 ......................446, 510, 517, 533 PDDRL ........................464, 510, 517, 533 PDIORL.......................445, 510, 517, 533 PECRH1 ......................448, 510, 517, 533 PECRH2 ......................448, 511, 517, 533 PECRL1.......................448, 511, 517, 533 PECRL2.......................448, 511, 517, 533 PEDRH ........................467, 510, 517, 533 PEDRL ........................467, 510, 517, 533 PEIORH.......................448, 510, 517, 533 PEIORL .......................448, 510, 517, 533 PIR...............................238, 514, 528, 537 PSR..............................241, 514, 529, 537 RBWAR ......................280, 514, 527, 537 RDFAR........................280, 514, 527, 537 RDLAR .......................264, 514, 526, 537 RFCR...........................244, 515, 530, 538 RFLR ...........................240, 514, 528, 537 RMCR .........................278, 514, 527, 537 RMFCR .......................275, 514, 526, 537 RTCNT ........................139, 513, 525, 536 RTCOR........................140, 513, 525, 536 RTCSR ........................137, 513, 525, 536 SCBRR ............................................... 346 SCBRR_0 ........................... 511, 519, 534 SCBRR_1 ........................... 512, 519, 534 SCBRR_2 ........................... 512, 520, 535 SCFCR................................................ 353 SCFCR_0............................ 512, 519, 534 SCFCR_1............................ 512, 520, 535 SCFCR_2............................ 512, 520, 535 SCFDR ............................................... 356 SCFDR_0 ........................... 512, 519, 534 SCFDR_1 ........................... 512, 520, 535 SCFDR_2 ........................... 512, 520, 535
Rev. 6.00 Jun. 12, 2007 Page 608 of 610 REJ09B0131-0600
SCFRDR ............................................. 330 SCFRDR_0 ......................... 512, 519, 534 SCFRDR_1 ......................... 512, 519, 535 SCFRDR_2 ......................... 512, 520, 535 SCFSR ................................................ 338 SCFSR_0 ............................ 512, 519, 534 SCFSR_1 ............................ 512, 519, 535 SCFSR_2 ............................ 512, 520, 535 SCFTDR ............................................. 331 SCFTDR_0 ......................... 511, 519, 534 SCFTDR_1 ......................... 512, 519, 535 SCFTDR_2 ......................... 512, 520, 535 SCLSR ................................................ 361 SCLSR_0 ............................ 512, 519, 534 SCLSR_1 ............................ 512, 520, 535 SCLSR_2 ............................ 512, 520, 535 SCRSR................................................ 330 SCSCR................................................ 334 SCSCR_0............................ 511, 519, 534 SCSCR_1............................ 512, 519, 534 SCSCR_2............................ 512, 520, 535 SCSMR ............................................... 331 SCSMR_0 ........................... 511, 519, 534 SCSMR_1 ........................... 512, 519, 534 SCSMR_2 ........................... 512, 520, 535 SCSPTR.............................................. 357 SCSPTR_0.......................... 512, 519, 534 SCSPTR_1.......................... 512, 520, 535 SCSPTR_2.......................... 512, 520, 535 SCTSR ................................................ 330 SDBPR................................................ 496 SDBSR................................................ 497 SDCR.......................... 136, 513, 524, 536 SDID ........................... 503, 511, 518, 534 SDIR ........................... 496, 511, 518, 534 STBCR........................ 218, 511, 518, 534 STBCR2...................... 219, 511, 519, 534 STBCR3...................... 220, 511, 518, 534 STBCR4...................... 221, 511, 518, 534 TBRAR ....................... 280, 514, 527, 537
TDFAR ....................... 281, 514, 527, 537 TDLAR....................... 264, 514, 525, 537 TFTR .......................... 275, 514, 526, 537 TLFRCR ..................... 244, 515, 530, 538 TPAUSER .................. 246, 515, 531, 538 TRIMD ....................... 282, 514, 527, 537 TROCR....................... 241, 514, 529, 537 TRSCER ..................... 273, 514, 526, 537 TSFRCR ..................... 243, 515, 530, 537 WTCNT ...................... 209, 511, 519, 534 WTCSR ...................... 209, 511, 519, 534 Register data format.................................. 24 Relationship between refresh requests and bus cycles................................................ 177 Reset ......................................................... 69 RISC-type ................................................. 25
System control instructions....................... 45 System registers ........................................ 22
T
TAP controller ........................................ 504 Transmit descriptor 0 (TD0) ................... 284 Transmit descriptor 1 (TD1) ................... 286 Transmit descriptor 2 (TD2) ................... 286 Transmitting and receiving serial data simultaneously (synchronous mode)....... 382 Transmitting serial data (asynchronous mode) .............................. 368 Transmitting serial data (synchronous mode)................................ 378 Trap instructions ....................................... 74 Types of exception handling and priority ...................................................... 65 Types of exceptions triggered by instructions................................................ 74 Types of power-down modes.................. 215
S
SCIF Initialization (Asynchronous Mode) ................................................................ 366 SCIF initialization (synchronous mode) . 376 SDRAM direct connection ..................... 153 SDRAM interface ................................... 153 Searching cache ........................................ 54 Self-refreshing ........................................ 176 Serial communication interface with FIFO (SCIF) ..................................................... 325 Shift instructions....................................... 44 Single read .............................................. 164 Single Write............................................ 167 Sleep mode ............................................. 222 Software standby mode........................... 223 Stack states after exception handling ends........................................................... 77 State transition .......................................... 47 Synchronous mode ................................. 375
U
U memory ................................................. 63 User break controller (UBC)................... 471 User break interrupt .................................. 97 User debugging interface (H-UDI) ......... 493
W
Wait between access cycles .................... 190 Watchdog timer (WDT) .......................... 207 Write access .............................................. 56 Write-back buffer...................................... 57
Rev. 6.00 Jun. 12, 2007 Page 609 of 610 REJ09B0131-0600
Rev. 6.00 Jun. 12, 2007 Page 610 of 610 REJ09B0131-0600
Renesas 32-Bit RISC Microcomputer Hardware Manual SH7618 Group
Publication Date: Rev.1.00, Feb. 18, 2004 Rev.6.00, Jun. 12, 2007 Published by: Sales Strategic Planning Div. Renesas Technology Corp. Edited by: Customer Support Department Global Strategic Communication Div. Renesas Solutions Corp.
2007. Renesas Technology Corp., All rights reserved. Printed in Japan.
Sales Strategic Planning Div.
Nippon Bldg., 2-6-2, Ohte-machi, Chiyoda-ku, Tokyo 100-0004, Japan
RENESAS SALES OFFICES
Refer to "http://www.renesas.com/en/network" for the latest and detailed information. Renesas Technology America, Inc. 450 Holger Way, San Jose, CA 95134-1368, U.S.A Tel: <1> (408) 382-7500, Fax: <1> (408) 382-7501 Renesas Technology Europe Limited Dukes Meadow, Millboard Road, Bourne End, Buckinghamshire, SL8 5FH, U.K. Tel: <44> (1628) 585-100, Fax: <44> (1628) 585-900 Renesas Technology (Shanghai) Co., Ltd. Unit 204, 205, AZIACenter, No.1233 Lujiazui Ring Rd, Pudong District, Shanghai, China 200120 Tel: <86> (21) 5877-1818, Fax: <86> (21) 6887-7898 Renesas Technology Hong Kong Ltd. 7th Floor, North Tower, World Finance Centre, Harbour City, 1 Canton Road, Tsimshatsui, Kowloon, Hong Kong Tel: <852> 2265-6688, Fax: <852> 2730-6071 Renesas Technology Taiwan Co., Ltd. 10th Floor, No.99, Fushing North Road, Taipei, Taiwan Tel: <886> (2) 2715-2888, Fax: <886> (2) 2713-2999 Renesas Technology Singapore Pte. Ltd. 1 Harbour Front Avenue, #06-10, Keppel Bay Tower, Singapore 098632 Tel: <65> 6213-0200, Fax: <65> 6278-8001 Renesas Technology Korea Co., Ltd. Kukje Center Bldg. 18th Fl., 191, 2-ka, Hangang-ro, Yongsan-ku, Seoul 140-702, Korea Tel: <82> (2) 796-3115, Fax: <82> (2) 796-2145
http://www.renesas.com
Renesas Technology Malaysia Sdn. Bhd Unit 906, Block B, Menara Amcorp, Amcorp Trade Centre, No.18, Jalan Persiaran Barat, 46050 Petaling Jaya, Selangor Darul Ehsan, Malaysia Tel: <603> 7955-9390, Fax: <603> 7955-9510
Colophon 6.0
SH7618 Group Hardware Manual

▲Up To Search▲

Price & Availability of HD6417618

	To Download HD6417618 Datasheet File
If you can't view the Datasheet, Please click here to try to view without PDF Reader .