CS5920

Part Number S5920 Revision 1.01 – November 28, 2005 S5920 PCI Product Data Sheet S5920 PCI PRODUCT AMCC Confidential and Proprietary DS1596 1 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book FEATURES • • • • • • • • • • • • • • • • • Full 132 Mbytes/sec Transfer Rate PCI Bus Operation to 33 MHz PCI Purposed 2.2 Compliant Target/Slave Device Add-On Bus up to 40 MHz Programmable Prefetch and Wait States 8/16/32-Bit Add-On Bus Four Definable Pass-Thru Regions Two 32-Byte Burstable FIFOs Active/Passive Add-On Bus Operation Mailbox Registers/w Byte Level Status Direct Mailbox Data Strobe/Int Pin Mailbox Read/Write Interrupts Direct PCI and Add-On Interrupt Pins Plug-N-Play Compatible Two-wire Serial Bus nvRAM Support Optional External BIOS capability 160-Pin PQFP with Green/RoHS compliant lead free option APPLICATIONS • • • • • • ISA Conversions Multimedia I/O Ports Data Storage CODEC5 General Purpose PCI Bus Interfacing ARCHITECTURAL OVERVIEW The AMCC S5920 was developed to provide the designer with a single multi-function device offering a flexible and easy way to connect to the PCI bus. By using the S5920, the designer eliminates the task of assuring PCI bus specification compliance and the necessity of understanding PCI bus timing require ments when interfacing a new application. The complex 33 MHz PCI bus signals are converted through the S5920 into an easy-to-use 8/16/32-bit user bus referred to as the user Add-On bus. The AddOn bus allows user add-on designs bus clock speed independent operation to 40 MHz. Figure 1. S5920 Block Diagram User Application S5920 PCI Configuration Registers ISA Card Design ADD-ON PassThru 32 Byte FIFO 32 Byte FIFO PCI PassThru AMCC ADD-ON Local Bus Interface Logic Mux/ Demux Active R/W Logic Buffers Serial Read/Write Control ISDN FDDI ATM Graphics/ MPEG/ Grabber PCI Local Bus 2.1 PCI Local Bus Interface Logic Mux/ Demux Data Buffers Serial Read/Write Control Pass-Thru Address Register Mailboxes/Status Operation/ Status Registers Audio Design I/O Controller Configuration Space Expansion BIOS Serial NVRAM Serial Bus AMCC Confidential and Proprietary DS1596 2 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book FEATURES .............................................................................................................................................................. 2 APPLICATIONS ...................................................................................................................................................... 2 ARCHITECTURAL OVERVIEW .............................................................................................................................. 2 S5920 Block Diagram ........................................................................................................................................ 2 VISION ................................................................................................................................................................... 10 CORPORATE OVERVIEW .................................................................................................................................... 10 AMCC Product Development Strategy ............................................................................................................ 10 Network Interface Products ............................................................................................................................. 10 AMCC Product Development Strategy ............................................................................................................ 10 Peripheral Component Interconnect (PCI) Bus Controllers ............................................................................. 11 Precision Clock and Timing Products .............................................................................................................. 11 Manufacturing Excellence ............................................................................................................................... 11 AMCC COMMITMENT TO QUALITY .................................................................................................................... 12 Reliability and Manufacturability—Designed In From The Start ...................................................................... 12 Quality Built In During Wafer Fabrication and Manufacturing .......................................................................... 12 Inspection, Audit and Reliability Confirmation ................................................................................................. 12 Continuous Quality Improvement Program ..................................................................................................... 13 ISO9001 REGISTRATION ..................................................................................................................................... 13 PRODUCT QUALIFICATIONS .............................................................................................................................. 14 MIL-STD-883 Method 5005 ............................................................................................................................. 14 “Qualification And Quality Conformance Procedures” ..................................................................................... 14 MIL-STD-883 Method 5010 ............................................................................................................................. 14 “Test Procedures For Custom Monolithic Microcircuits” .................................................................................. 14 Qualification Method 5005 VS. 5010 ............................................................................................................... 14 Generic Data ................................................................................................................................................... 14 AMCC Product Assurance Product Flow Detail .............................................................................................. 15 AMCC’S RELIABILITY VIGIL ............................................................................................................................... 17 New/Changed Wafer Processes and Material Qualifications .......................................................................... 17 Wafer Process and Design .............................................................................................................................. 17 Package and Related Materials ....................................................................................................................... 17 In-Process Quality Monitors ............................................................................................................................ 17 Periodic Operating Life and Environmental Testing ........................................................................................ 17 Final Measure and Assurance of Quality ......................................................................................................... 17 RADIATION HARDNESS ...................................................................................................................................... 17 ATM LAN and 100VG AnyLAN Products ........................................................................................................ 18 Fibre Channel/Gigabit Ethernet Products ........................................................................................................ 18 PCI Products ................................................................................................................................................... 19 SONET/SDH/ATM Products ............................................................................................................................ 19 HIPPI Products ................................................................................................................................................ 19 Crosspoint Switch Products ............................................................................................................................. 20 Precision Clocking Products ............................................................................................................................ 20 SONET/SDH/ATM Products (continued) ......................................................................................................... 20 ASIC Standard Cell Products .......................................................................................................................... 21 Clock Generator and Synthesizer Products .................................................................................................... 21 ASIC Logic Array Products .............................................................................................................................. 22 ARCHITECTURAL OVERVIEW S5920 ................................................................................................................. 23 AMCC Confidential and Proprietary DS1596 3 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book S5920 REGISTER ARCHITECTURE .................................................................................................................... 23 PCI Configuration Registers ............................................................................................................................ 23 PCI Bus Accessible Registers ......................................................................................................................... 23 Add-On Bus Accessible Registers ................................................................................................................... 23 SERIAL NON-VOLATILE INTERFACE ................................................................................................................ 23 S5920 Pinout ................................................................................................................................................... 24 MAILBOX OPERATION ........................................................................................................................................ 25 PASS-THRU OPERATION .................................................................................................................................... 25 Mailbox Block Diagram .................................................................................................................................... 25 Pass-Thru Block Diagram ................................................................................................................................ 26 S5920 Pin Assignment .................................................................................................................................... 28 PCI Bus Address and Data Signal ................................................................................................................... 29 PCI Bus System Signals .................................................................................................................................. 30 PCI Bus Data Transfer Control Signals ........................................................................................................... 30 PCI Bus Error Reporting Signals ..................................................................................................................... 30 ADD-ON BUS AND S5920 CONTROL SIGNALS ................................................................................................ 31 Serial nvRAM Interface Signals ....................................................................................................................... 31 Direct Mailbox Access Signals ........................................................................................................................ 31 USER ADD-ON BUS PIN DESCRIPTIONS .......................................................................................................... 32 Pass-Thru Data Channel Pins ......................................................................................................................... 32 S5920 Add-On Bus Register Access Pins ....................................................................................................... 33 Add-On Bus General Pins ............................................................................................................................... 34 Configuration Registers ................................................................................................................................... 35 VENDOR IDENTIFICATION REGISTER (VID) ..................................................................................................... 37 Vendor Identification Register ......................................................................................................................... 37 DEVICE IDENTIFICATION REGISTER (DID) ....................................................................................................... 38 Device Identification Register .......................................................................................................................... 38 PCI COMMAND REGISTER (PCICMD) ................................................................................................................ 39 PCI Command Register ................................................................................................................................... 39 PCI STATUS REGISTER (PCISTS) ...................................................................................................................... 41 PCI Status Register ......................................................................................................................................... 41 REVISION IDENTIFICATION REGISTER (RID) ................................................................................................... 43 Revision Identification Register ....................................................................................................................... 43 CLASS CODE REGISTER (CLCD) ....................................................................................................................... 44 Class Code Register ........................................................................................................................................ 44 Defined Base Class Codes .............................................................................................................................. 45 Base Class Code 00h: Early, Pre-2.0 Specification Devices ........................................................................... 45 Base Class Code 01h: Mass Storage Controllers ........................................................................................... 45 Base Class Code 02h: Network Controllers .................................................................................................... 45 Base Class Code 03h: Display Controllers ...................................................................................................... 46 Base Class Code 04h: Multimedia Devices ..................................................................................................... 46 Base Class Code 05h: Memory Controllers .................................................................................................... 46 Base Class Code 06h: Bridge Devices ............................................................................................................ 47 Base Class Code 07h: Simple Communications Controllers ........................................................................... 47 Base Class Code 08h: Base System Peripherals ........................................................................................... 47 Base Class Code 09h: Input Devices .............................................................................................................. 48 Base Class Code 0Ah: Docking Stations ........................................................................................................ 48 AMCC Confidential and Proprietary DS1596 4 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Base Class Code 0Bh: Processors ................................................................................................................. 48 Base Class Code 0Ch: Serial Bus Controllers ................................................................................................ 48 CACHE LINE SIZE REGISTER (CALN) ............................................................................................................... 49 Cache Line Size Register ................................................................................................................................ 49 LATENCY TIMER REGISTER (LAT) .................................................................................................................... 50 Latency Timer Register ................................................................................................................................... 50 HEADER TYPE REGISTER (HDR) ....................................................................................................................... 51 Header Type Register ..................................................................................................................................... 51 BUILT-IN SELF-TEST REGISTER (BIST) ............................................................................................................ 52 Built-In Self-Test Register ................................................................................................................................ 52 BASE ADDRESS REGISTER (BADR) ................................................................................................................. 53 Determining Base Address Size ...................................................................................................................... 53 Assigning the Base Address ............................................................................................................................ 53 Base Address Register - Memory .................................................................................................................... 54 Base Address Register - I/O ............................................................................................................................ 55 Base Address Register Response (Memory Assigned) to All-Ones Write Operation ..................................... 56 Read Response (I/O Assigned) to an All-Ones Write Operation to a Base Address Register ........................ 57 SUBSYSTEM VENDOR IDENTIFICATION REGISTER (SVID) ........................................................................... 58 Subsystem Vendor Identification Register ....................................................................................................... 58 SUBSYSTEM ID REGISTER (SID) ....................................................................................................................... 59 Subsystem Identification Register ................................................................................................................... 59 EXPANSION ROM BASE ADDRESS REGISTER (XROM) ................................................................................. 60 Expansion ROM Base Address Register ......................................................................................................... 60 Read Response to Expansion ROM Base Address Register (after all ones written) ...................................... 61 INTERRUPT LINE REGISTER (INTLN) ................................................................................................................ 62 Interrupt Line Register ..................................................................................................................................... 62 INTERRUPT PIN REGISTER (INTPIN) ................................................................................................................ 63 Interrupt Pin Register ....................................................................................................................................... 63 MINIMUM GRANT REGISTER (MINGNT) ............................................................................................................ 64 Minimum Grant Register .................................................................................................................................. 64 MAXIMUM LATENCY REGISTER (MAXLAT) ...................................................................................................... 65 Maximum Latency Register ............................................................................................................................. 65 OPERATION REGISTERS .................................................................................................................................... 66 PCI BUS OPERATION REGISTERS .................................................................................................................... 66 Operation Registers - PCI Bus ........................................................................................................................ 66 OUTGOING MAILBOX REGISTER (OMB) ........................................................................................................... 67 Outgoing Mailbox ............................................................................................................................................. 67 PCI INCOMING MAILBOX REGISTER (IMB) ....................................................................................................... 68 Incoming Mailbox ............................................................................................................................................. 68 PCI MAILBOX EMPTY/FULL STATUS REGISTER (MBEF) ................................................................................ 69 Mailbox Empty/Full Status Register (MBEF) ................................................................................................... 69 Mailbox Empty/Full Status Register ................................................................................................................. 69 PCI INTERRUPT CONTROL/STATUS REGISTER (INTCSR) ............................................................................. 70 Interrupt Control Status Register ..................................................................................................................... 70 Interrupt Control Status Register ..................................................................................................................... 71 PCI RESET CONTROL REGISTER (RCR) ........................................................................................................... 72 AMCC Confidential and Proprietary DS1596 5 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book FIFO Control/Status Register .......................................................................................................................... 72 Reset Control Register .................................................................................................................................... 73 PCI PASS-THRU CONFIGURATION REGISTER (PTCR) ................................................................................... 74 Pass-Thru Configuration Register ................................................................................................................... 74 ......................................................................................................................................................................... 75 ADD-ON BUS OPERATION REGISTERS ............................................................................................................ 76 Operation Registers - Add-On Interface .......................................................................................................... 76 ADD-ON INCOMING MAILBOX REGISTER (AIMB) ............................................................................................ 77 ADD-ON OUTGOING MAILBOX REGISTER (AOMB) ......................................................................................... 77 ADD-ON PASS-THRU ADDRESS REGISTER (APTA) ........................................................................................ 77 ADD-ON PASS-THRU DATA REGISTER (APTD) ............................................................................................... 77 ADD-ON MAILBOX EMPTY/FULL STATUS REGISTER (AMBEF) ..................................................................... 78 Mailbox Empty/Full Status Register ................................................................................................................. 78 Mailbox Empty/Full Status Register ................................................................................................................. 79 ADD-ON INTERRUPT CONTROL/STATUS REGISTER (AINT) .......................................................................... 80 Add-On Interrupt Control Status Register ........................................................................................................ 80 Interrupt Control Status Register ..................................................................................................................... 81 ADD-ON RESET CONTROL REGISTER (ARCR) ................................................................................................ 82 Add-On General Control/Status Register ........................................................................................................ 82 Reset General Control/Status Register ........................................................................................................... 83 ADD-ON PASS-THRU CONFIGURATION REGISTER (APTCR) ........................................................................ 84 Pass-Thru Configuration Register ................................................................................................................... 84 Pass-thru Configuration Register .................................................................................................................... 85 INTRODUCTION ................................................................................................................................................... 86 PCI RESET ............................................................................................................................................................ 86 LOADING THE SERIAL NV MEMORY ................................................................................................................. 86 Valid External Boot Memory Contents ............................................................................................................. 87 S5920 to nvRAM Interface .............................................................................................................................. 88 Serial Interface Definition of Start and Stop .................................................................................................... 88 Serial Interface Clock Data Relationship ......................................................................................................... 88 Serial Interface Byte Access-Write .................................................................................................................. 89 Serial Interface Byte Access-Read .................................................................................................................. 89 Serial Byte Access- Sequential Read .............................................................................................................. 89 NON-VOLATILE MEMORY INTERFACE ............................................................................................................. 90 NVRAM READ/WRITE DESCRIPTION ................................................................................................................. 90 PCI BUS CONFIGURATION CYCLES .................................................................................................................. 93 PCI AD Bus Definition Type 0 Configuration Access ...................................................................................... 93 Type 0 Configuration Read Cycles .................................................................................................................. 93 Type 0 Configuration Write Cycles .................................................................................................................. 94 EXPANSION BIOS ROMS .................................................................................................................................... 94 PC Compatible Expansion ROM ..................................................................................................................... 94 PCI Data Structure .......................................................................................................................................... 96 PCI BUS INTERFACE ........................................................................................................................................... 97 PCI BUS TRANSACTIONS ................................................................................................................................... 97 PCI Bus Commands ........................................................................................................................................ 97 PCI BURST TRANSFERS ..................................................................................................................................... 98 AMCC Confidential and Proprietary DS1596 6 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book PCI READ TRANSFERS ....................................................................................................................................... 98 Single Data Phase PCI Bus Read of S5920 Registers or Expansion ROM .................................................... 98 PCI WRITE TRANSFERS ..................................................................................................................................... 99 Burst PCI Bus Read Attempt to S5920 Registers or Expansion ROM ............................................................ 99 Burst PCI Bus Write of S5920 Registers ......................................................................................................... 99 Target-Initiated Termination ............................................................................................................................ 99 Target Disconnects .......................................................................................................................................... 99 Target Requested Retries ............................................................................................................................. 100 Figure 4a. Target Disconnect Example 1 ...................................................................................................... 100 Figure 4b. Target Disconnect Example 2 ...................................................................................................... 100 Target Aborts ................................................................................................................................................. 100 Target Latency ............................................................................................................................................... 101 Target Locking ............................................................................................................................................... 101 PCI Bus Access Latency Components .......................................................................................................... 101 Figure 5. Target-Initiated Retry ...................................................................................................................... 101 Figure 6. Engaging the LOCK# Signal .......................................................................................................... 101 Target Termination Type ............................................................................................................................... 102 PCI BUS INTERRUPTS ...................................................................................................................................... 103 PCI BUS PARITY ERRORS ................................................................................................................................ 103 Access to a Locked Target by its Owner ....................................................................................................... 104 Access Attempt to a Locked Target ............................................................................................................... 104 Error Reporting Signal ................................................................................................................................... 104 MAILBOX OVERVIEW ........................................................................................................................................ 105 FUNCTIONAL DESCRIPTION ............................................................................................................................ 105 Figure 1. PCI to Add-On Mailbox Register .................................................................................................... 105 Add-On to PCI Mailbox Register ................................................................................................................... 106 Mailbox Empty/Full Conditions ...................................................................................................................... 106 Mailbox Interrupts .......................................................................................................................................... 106 Add-On Outgoing Mailbox, Byte 3 Access .................................................................................................... 107 BUS INTERFACE ................................................................................................................................................ 107 PCI Bus Interface .......................................................................................................................................... 107 Add-On Bus Interface .................................................................................................................................... 107 8-Bit and 16-Bit Add-On Interfaces ................................................................................................................ 108 CONFIGURATION ............................................................................................................................................... 108 Mailbox Status ............................................................................................................................................... 108 Input/Output Mode (MDMODE=0) ................................................................................................................. 108 Input Mode (MDMODE=1) ............................................................................................................................. 108 Mailbox Interrupts .......................................................................................................................................... 110 Servicing a PCI Mailbox Interrupt (INTA# asserted): ..................................................................................... 111 ADD-ON LOCAL BUS INTERFACE ................................................................................................................... 113 ADD-ON INTERFACE SIGNALS ........................................................................................................................ 113 SYSTEM SIGNALS ............................................................................................................................................. 113 ADD-ON S5920 REGISTER ACCESSES ........................................................................................................... 113 Register Access Signals ................................................................................................................................ 113 S5920 General Register Accesses ................................................................................................................ 113 Read Operation Register ............................................................................................................................... 114 Write Operation Register ............................................................................................................................... 114 AMCC Confidential and Proprietary DS1596 7 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book S5920 16-bit Mode Register Accesses ......................................................................................................... 114 16 Bit Mode Operation Register DWORD Write/Read .................................................................................. 115 MAILBOX OVERVIEW ........................................................................................................................................ 116 PASS-THRU OVERVIEW .................................................................................................................................... 116 WRITE FIFO OVERVIEW .................................................................................................................................... 116 FUNCTIONAL DESCRIPTION ............................................................................................................................ 116 Pass-Thru Transfers ...................................................................................................................................... 116 Pass-Thru Status/Control Signals ................................................................................................................. 117 BUS INTERFACE ................................................................................................................................................ 117 PCI Bus Interface .......................................................................................................................................... 117 PCI Pass-Thru Single Cycle Accesses .......................................................................................................... 118 PCI Pass-Thru Burst Accesses ..................................................................................................................... 118 PCI Disconnect Conditions ............................................................................................................................ 118 PCI Write Disconnect .................................................................................................................................... 119 PCI Read Disconnect .................................................................................................................................... 119 S5920 PASSIVE MODE OPERATION ................................................................................................................ 120 Single-Cycle PCI to Pass-Thru Write ............................................................................................................ 120 PCI To Add-On Passive Write ....................................................................................................................... 121 PCI To Add-On Passive Write w/Pass-Thru Address .................................................................................... 121 Single-Cycle PCI to Pass-Thru Read ............................................................................................................ 121 PCI To Add-On Passive Read ....................................................................................................................... 122 PCI to Pass-Thru Burst Writes ...................................................................................................................... 122 PCI to Add-On Passive Burst Write ............................................................................................................... 123 PCI to Add-On Passive Burst Write Using PTRDY# to assert Wait-States ................................................... 125 PCI to Add-On Passive Burst Read Access .................................................................................................. 125 Pass-Thru Burst Reads ................................................................................................................................. 126 PCI to Add-On Passive Burst Read ............................................................................................................... 127 Using PTRDY# to assert Wait-States ............................................................................................................ 128 8-Bit and 16-Bit Pass-Thru Add-On Bus Interface in Passive Mode ............................................................. 129 Byte Lane Steering for PCI Write (Add-On Read) ......................................................................................... 130 Byte Lane Steering for PCI Read (Add-On Write) ......................................................................................... 130 PCI to Add-On Passive Write to an 8-bit ....................................................................................................... 131 Endian Conversion ........................................................................................................................................ 132 Active Operation ............................................................................................................................................ 133 Showing Big Endian Conversion for 32-bit .................................................................................................... 133 PCI to Add-On Passive Read to an 16-bit Add-On Device ............................................................................ 133 Big Endian conversion for a 16-bit bus. The S5920 drives D[15:0] only ....................................................... 134 Big Endian conversion for an 8-bit bus. The S5920 drives D[7:0] only ......................................................... 134 Active mode PCI Read (Zero Programmed Wait States) with PTADR# ........................................................ 134 Active Mode PCI Read without PTADR# ....................................................................................................... 134 Active Mode PCI Write without PTADR# ....................................................................................................... 135 Active mode Programmable Wait States ....................................................................................................... 135 PTRDY#/PTWAIT# ........................................................................................................................................ 135 Active Mode PCI Write with Add-On Initiated Wait States Using PTWAIT# .................................................. 135 Active Mode 32-Bit PCI Write ........................................................................................................................ 136 DXFR# ........................................................................................................................................................... 136 Active Mode Figures and Descriptions .......................................................................................................... 136 AMCC Confidential and Proprietary DS1596 8 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Active mode Burst cycles .............................................................................................................................. 136 Active Mode 32-Bit PCI Write w/PTWAIT# .................................................................................................... 137 Clock by Clock description of Figure 16 ........................................................................................................ 137 Clock by Clock description of Figure 17 ........................................................................................................ 138 Active Mode PCI Write Showing a One Wait State Programmed Delay ....................................................... 138 16-Bit Active Mode PCI Read w/ Programmed Wait States .......................................................................... 138 Active Mode with 16/8-bit data buses ............................................................................................................ 139 Active Mode PCI Read w/ Programmed Wait States .................................................................................... 139 Active Mode PCI Read .................................................................................................................................. 140 Active Mode PCI Write .................................................................................................................................. 140 CONFIGURATION ............................................................................................................................................... 140 S5920 Base Address Register Definition ...................................................................................................... 141 8-Bit Active Mode PCI Write .......................................................................................................................... 141 Creating a Pass-Thru Region ........................................................................................................................ 141 Accessing a Pass-Thru Region ..................................................................................................................... 142 Special Programming Features ..................................................................................................................... 142 ABSOLUTE MAXIMUM STRESS RATINGS ...................................................................................................... 144 Absolute Maximum Stress Ratings ................................................................................................................ 144 Operating Conditions ..................................................................................................................................... 144 PCI Signal DC Characteristics ....................................................................................................................... 145 PCI Signal DC Characteristics (V CC = 5.0V 5%, 0 0 C to 70 0 C, 50 pF load on outputs) .......................... 145 Add-On Operating Characteristics (V CC = 5.0V 5%, 0 0 C to 70 0 C, 50 pF load on outputs) .................... 146 nvRAM Memory Interface Signals ................................................................................................................. 146 TIMING SPECIFICATION .................................................................................................................................... 147 PCI Clock Specification ................................................................................................................................. 147 Functional Operation Range (VCC = 5.0V +/- 5%, 0O C to 70 OC, 50 pF load on outputs for MAX, 0 pF load for MIN) .................. 147 PCI Clock Timing ........................................................................................................................................... 147 PCI Signal Output Timing .............................................................................................................................. 148 PCI Signal Input Timing ................................................................................................................................. 148 Add-On Timings, Functional Operation Range (VCC = 5.0 V ± 5%, 0°C to 7 0°C, 50 pF load on outputs for MAX, 0 pF load for MIN) ........................... 149 Add-On Clock Timing .................................................................................................................................... 151 Pass-Thru Clock Relationship to PCI Clock .................................................................................................. 151 PTADR Timing ............................................................................................................................................... 151 Passive Mode Pass-Thru Operation .............................................................................................................. 152 Add-On Timings Functional Operation Range (VCC = 5.0 V ± 5%, 0°C to 70°C, 50 pF load on outputs for MAX, 0 pF load for MIN) ........................................................................................................................................................ 153 Active Mode Pass-Thru Write Operation ....................................................................................................... 154 Mailbox Data .................................................................................................................................................. 155 Mailbox Data .................................................................................................................................................. 155 Mailbox Timings Functional Operation Range (VCC = 5.0 V ± 5%, 0°C to 70°C, 50 pF load on outputs) .......................... 155 S5920 Pinout and Pin Assignment ................................................................................................................ 156 S5920 – 160 PQFP PACKAGE MARKING DRAWING (TOP VIEW) ................................................................. 157 S5920 – 160 PQFP Package Marking Drawing (Top View) .......................................................................... 157 160 PQFP (28 x 28 x 3.37 mm) - Plastic Quad Flat Package ....................................................................... 158 Ordering Information ...................................................................................................................................... 160 AMCC Confidential and Proprietary DS1596 9 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book VISION It is AMCC’s vision to be the premier supplier of silicon for high bandwidth connectivity solutions for the worldwide networking infrastructure. AMCC Product Development Strategy AMCC’s product development strategy utilizes the company’s expertise in the telecommunications, data communications, ATE, computer and military markets. Initially ASSPs are defined based on key industry standards. Then, as additional applications are identified, derivative products based on the “cores” of the original devices are introduced. These “cores” are then made available for high volume proprietary customer specific designs. This is all made possible by our emphasis on using ASIC techniques and methodologies to develop the original products. Network Interface Products High performance network interface encompasses a wide range of applications, all requiring data transmission rates from >100 Mbit per second to over 9.6 Gbit per second. These applications include Local Area Networks (LAN), Hubs, Routers and Switches, Wide Area Networks (WAN) based on SONET standards, RAID-based storage systems, serial backplane for high performance switches and digital video broad casting systems. CORPORATE OVERVIEW AMCC defines, develops, manufactures and markets application specific standard products (ASSPs) and customer specific integrated circuits (CSICs) for high speed, high performance network interface applications. Utilizing CMOS, BiCMOS and proprietary MicroPower Bipolar technology, AMCC provides precision circuits and interface solutions for Gigabit Ethernet, ATM, SONET, Fibre Channel and PCI markets. Since 1979, AMCC has designed and produced five generations of semicustom bipolar ECL logic arrays and two generations of BiCMOS logic arrays. AMCC expertise includes its mixed ECL/TTL interface, phase-locked loop (PLL), precision vernier, skew control, high-speed VCO and controlled edge rate TTL outputs. Figure 2. AMCC Product Development Strategy Consumer Telecom Auto Standard Industrial ASIC Methodology Military Customer Specific Derivative Datacom ATE AMCC Confidential and Proprietary DS1596 10 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book AMCC interface circuits, transceiver chips and switches are designed to implement emerging network technologies such as the ANSI Fibre Channel and High Performance Interface (HIPPI) standards, the ITU SONET telecommunications standard, the ATM Forum LAN standard and IEEE Gigabit Ethernet standard. Jitter, speed, power and size are critical design issues for all these circuits. AMCC’s devices are based on its unique Bipolar process which has noise isolation characteristics that enable best-in-class jitter performances. The inherent physical structure of the company’s process makes 1 to 3 GigaHertz (GHz) data rates possible at low power. Consequently, the low device power consumption of AMCC’s products helps to minimize the cost and size of packaging. specific customer needs for high performance clock generation and distribution, clock synchronization and de-skewing, frequency synthesis, and pulse shaping applications. Offerings include low EMI, low skew clock drivers and low jitter clock generators for high performance server, work-station and RAMBUS™based applications. Manufacturing Excellence AMCC manufactures its own Bipolar and BiCMOS wafers using proven processes and utilizes foundry relationships for access to CMOS capability in its world class fab located in San Diego. AMCC is proud of its best-in-class status for lowest defect densities for like sized fabs. This allows AMCC to provide a continuous, predictable supply of product to its customers. The Company follows a “semi-fabbed” manufacturing strategy and has CMOS and BiCMOS foundry rela tionships in place with major domestic and international semiconductor partners that provide for significant additional production capacity. Wafer purchases from strategic foundry partners both expand capacity and provide alternate sources. Additional high-volume assembly and test facilities are located offshore. AMCC’s “quick turn, semi-fabbed” manufacturing approach blends together the strengths of both the “fabbed” and “fabless” semiconductor strategies. Fabbed advantages include the security of total inhouse control and time to market. Fabless advantages include multiple sourcing and allows the company to focus investment on new high performance products. Peripheral Component Interconnect (PCI) Bus Controllers Increasing bandwidth of high speed networks creates a bottleneck at the desktop. One of the causes of this problem is the latency associated with connection to high speed peripheral equipment over LANs and WANs. The 132 Megabyte per second backplane PCI bus helps break the bottleneck. AMCC has developed the industry’s first line of general purpose master/slave controllers for the PCI bus. These circuits provide a high performance single-chip interface for add-on boards and adapter cards for industrial, graphics, video and communications markets. Precision Clock and Timing Products AMCC provides a line of high precision clock and timing standard products for exacting system designs. AMCC has also tailored clock and timing devices to AMCC Confidential and Proprietary DS1596 11 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book AMCC COMMITMENT TO QUALITY AMCC is committed to achieving the highest quality and reliability level in the integrated circuit products we provide. Every year for over a decade we have established industry-leading reliability and outgoing quality targets and then exceeded them. The quality and reliability philosophy at AMCC starts with the premise that for AMCC to continue to excel and be the premier supplier to our customers, the quality expectations of customers must be consistently met or exceeded. Our team operating philosophy is to: 1. design in manufacturability and reliability during the new product development phase (plan); 2. build in quality at all manufacturing steps (do); 3. execute thorough product inspections, internal audits and reliability confirmation (check); 4. incorporate feedback from internal and external sources into continuous quality improvement programs (act). Customer Documentation Design/Manufacturing QA gates and subsequent feedback ensures quality confirmation of AMCC’s final product in a continuous improvement program. Inspection, Audit and Reliability Confirmation AMCC has strategically placed In-Process Quality Control (IPQC) gates, internal process/area audits and lot/time specific reliability monitors to verify perfor mance against customer requirements and internal design/manufacturing process capabilities. Metrics generated by these activities are intended to provide continuous improvement feedback data for review and action as driven by senior management. • • • • • • • • Die visual and precap gate Final outgoing inspection gate Modified MIL-STD-105D sampling program Lot specific group A and B testing Ongoing reliability monitors SPC/Data metric review of key subcontractors Visual/mechanical and electrical outgoing indices and PPM goals Cost-of-quality pareto analysis Reliability and Manufacturability—Designed In From The Start Reliability and manufacturability is designed in up front through a team infrastructure which focuses on active participation by Design, Manufacturing and Reliability Engineering throughout all phases of the design process. This includes extensive design verification through computer modeling and design validation by product characterization and application simulation. Final team design review and production readiness approval is required prior to release of products to production. Quality Built In During Wafer Fabrication and Manufacturing AMCC’s manufacturing and quality teams employ documented operating procedures, work instructions, inprocess inspections and SPC methodology to provide assurance of continued process control and compli ance to specification. AMCC Confidential and Proprietary DS1596 12 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Continuous Quality Improvement Program • • • • • Corporate-wide commitment driven by the Executive Staff A program plan that is flexible enough to comprehend dynamic customer inputs Statistical tools in place for analysis and action planning Weekly and monthly review meetings to share performance data Self examination consistent with elements in ISO9001 and the Malcolm Baldridge National Quality Award AMCC QUALITY SYSTEM The Quality System had been modeled after the stringent military requirements of MIL-I-45208, MIL-Q-9858 and MIL-I-38535 Appendix A. Heading into the 21st century, AMCC has now modified its Quality System to also align with ISO9001. This has strengthened the closed loop improvement cycle by tying internal audits with corrective/preventative action though continuous management review. AMCC’s Quality System has the following components integrated throughout the factory to meet or exceed the above requirements. • • • • • • • • • • • • • • • • • • Quality Organization Quality Planning Management Review Contract Review Design Control Document and Data Control Purchasing Supplier Selection and Control Control of Customer Supplied Materials Product Identification and Traceability Operating Procedures Work Instructions Inspection and Test Inspection, Measurement and Test Equipment Calibration Inspection Status System Control of Nonconforming Material Corrective and Preventive Action ESD Safe Handling, Storage, Packaging, Preservation and Delivery Methods Records Retention and Maintenance Internal Process/Area Auditing System Training/Certification SPC and Statistical Techniques Failure Analysis AMCC Quality Philosophy ISO9001 AMCC Quality Manual AMCC QA Practices Manufacturing Materials Inspection and Test Design Customer Conformance Rqmts • • • • • Superior Product Quality ISO9001 REGISTRATION Based on the restructure of the Quality System to ISO9001 requirements and successful completion of internal and third-party audits, AMCC was ISO registered on July 29, 1996. Bi-yearly surveillance audits have been successfully passed as well. Please contact the factory for further details and schedule updates. AMCC Confidential and Proprietary DS1596 13 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book PRODUCT QUALIFICATIONS A qualification is a sequence of tests in which all parameters, including the reliability of the device are tested. It is this sequence of tests which initially qualifies the part to be released for production. Thorough reliability testing is performed on new product and package families in order to ensure the expectations of our customers are met. These tests include environmental, mechanical and life testing performed in accordance with Military Standards, industrial accepted methods and AMCC Test Procedures. Contact the factory for specific details regarding your selected product/package combination. AMCC provides MIL-STD-883 Methods 5005 and 5010 testing for our military customers on contract as well as MIL-H-38534 quality conformance screening for hybrid customers. Group D verifies the material integrity and the reliability of the package. Group E demonstrates the radiation hardness capability of the device. Performed on a generic basis by device type or as required for an application. MIL-STD-883 Method 5010 “Test Procedures For Custom Monolithic Microcircuits” This method establishes screening and quality con formance procedures for the testing of custom and semicustom monolithic semiconductors to verify Class B or Class S quality and reliability levels. Testing is performed in conjunction with other documentation such as MIL-I-38535 and an applicable detail specification. It establishes the design, material, performance, control, and documentation requirements needed to achieve prescribed levels of device quality and reliability. AMCC can support qualification using this method. Until August of 1983, the qualification most commonly used was Method 5005. Since that time, the newer revision of MIL-STD-883 includes Method 5010, which is better suited for semicustom devices (logic arrays included). Either qualification is adequate, but it is desirable to use the 5010 qualification procedure in qualifying custom or semicustom devices. MIL-STD-883 Method 5005 “Qualification And Quality Conformance Procedures” Method 5005 establishes qualification and quality-conformance inspection procedures for semi-conductors to ensure that the quality of devices and lot conform with the requirements of the applicable procurement document. The full requirements of Group A, B, C, D, and E test and inspections are intended for use in initial device qualification—or requalification in the event of product or process change—and in periodic testing for retaining qualification. Qualification Method 5005 VS. 5010 The primary difference between the two methods is in the Group D test. Method 5005 uses electrically-good devices, where method 5010 uses electrical rejects and package-only parts for environmental tests. In addition, Method 5010 is designed for smaller production releases (i.e., 2000 devices/year) while Method 5005 is designed for large production releases. Group A consists of electrical tests performed on an inspection lot which has already passed the 100% screening requirements. After a lot has passed the 100% screen tests, a random sample of parts is selected from the total population of devices to form the inspection lot. The inspection lot is then subjected to these Group A electrical tests. Group B inspection tests are used to monitor the fabrication and assembly processes performed on each inspection lot. Group C consists of a 1000-hour life test conducted to verify die integrity. Generic Data Under the provision of MIL-I-38535, a customer can elect to qualify using generic data (similar device/family). However, the provisions of the applicable contract should be reviewed. In most cases generic data will satisfy full qualification requirements. Since many of the qualifications at AMCC are ongoing, generic data may be available for this purpose. AMCC Confidential and Proprietary DS1596 14 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book AMCC Product Assurance Product Flow Detail Design For Manufacturability And Reliability • • Incoming Inspection • • • • • Wafer Fabrication Component Selection • Definition — Specification • Supplier: Selection — Qualification — Approval • Qualified Vendors List (QVL) Acceptance Documents and Operating Procedures • Purchase Order • Component/Material Specifications • Product Assurance and General Procedures • QVL • Sample Inspection of All Direct Materials Class 10 Clean Room — FED-STD-209 Measurements in Adherence with MIL-STD977 SPC: • In-process Monitors • PCM Electricals SEM Inspection on All Military Lots QA Audits • CV Plots —Weekly (minimum) • DI Water —Weekly • Particle Counts • Bacteria Count • Airborne Particle Count —Weekly Wafer Electrical Test Wafer Stores AMCC Confidential and Proprietary DS1596 15 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book AMCC Product Assurance Product Flow Detail Assembly Issue • IPQC Audit • Manufacturing Procedures Assembly and Environmental Screening Hermetic Military Flow • Method 5004 1 • IPQC • Customer Source Inspection (CSI) • Assembly Final Inspection • Methods 5004 & 5005 1 Hermetic Commercial Flow • Method 5004 (No Centrifuge) • IPQC • Assembly Final Inspection Plastic Flow • Precap Visual • Die Shear SPC • Wire Bond SPC • Mold/External Visual • Lead Trim and Coplanatary Inspection Pre-Electrical and Burn-In Final Electrical Test • Methods 5004 & 5005 1 • IPQC • Group A • 25°C — 100% • 0°C & 70°C — AQL=0.25% Sample Group B • Method 5005 1 • IPQC • Final Inspection • IPQC • Final Inspection • Coplanatary Check Group C & D • Method 5005 1 CSI • Per Order Packaging for Shipment • IPQC • Outgoing Inspection • IPQC • Outgoing Inspection Ship 1. MIL-H-38534 or MIL-STD-883 method 5010 also available. AMCC Confidential and Proprietary DS1596 16 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book AMCC’S RELIABILITY VIGIL AMCC’s internal reliability vigil consists of three phases: • • • New/changed processes and material qualifications In-process Quality monitors Periodic operating life and environmental testing Periodic Operating Life and Environmental Testing • • • • Performed on a product from each process family quarterly. 1000 hour operating life test (minimum), Method 5005, Group C. Temperature cycling per Method 1010, 100 cycles, condition C: –65°C/150°C Environmental testing per AMCC standard test procedures. Consult factory for further details. New/Changed Wafer Processes and Material Qualifications Final Measure and Assurance of Quality In order to initially release a device to production a standard set of MIL-STD-883 tests must be completed successfully. These tests include: Wafer Process and Design • • • • Operating Life Method 1005 ESD Characterization Method 3015 Wire Bond Pull Method 2011 Thermal Shock or Method 1011 or 1010 Temperature Cycling Selected Subgroups of MIL-STD-883, Method 5005, Group B and D The cost of defects depends on when the failure occurs. For example, costs rise significantly as undetected defective ICs are integrated into systems. High quality parts cut costs substantially, and the extra quality built into every AMCC device means added value to our customers. To achieve maximum quality, AMCC employs 100% testing of all devices, followed by stringent QA sampling. AMCC performs QA sampling measurements at full specification temperature, both DC and AC, to achieve the tightest AQLs in the industry. Package and Related Materials • RADIATION HARDNESS High energy radiation can cause structural changes in the silicon and silicon dioxide crystal lattice by displacing atoms from their normal crystal sites. These changes can be responsible for increased junction leakage, degraded transistor current gain (b), and increased parasitic Si/SiO2 interface leakage currents. The damage is generally induced by neutrons, X-rays, and gamma rays. The effects of the damage induced by this radiation can change both AC and DC parameters, affect functional performance, and, in severe cases, destroy the device. Certain of AMCC’s high performance products are inherently radiation resistant. The radiation resistance of AMCC IC’s is the result of the small geometries, the structure of the fabrication process itself, and the use of ECL logic within the device. Contact your AMCC representative regarding radiation resistance characteristics associated with a specific product. AMCC adheres to MIL-I-38535 with regards to changes. If changes to production released devices are determined to be major, the appropriate qualification testing must be successfully completed prior to change approval. In-Process Quality Monitors • • • • • • • • CV plots Airborne particle count Bacteria, particle count, and resistivity on DI water ESD work stations and procedures In-line testing of process gases Temperature and humidity control SPC in wafer fabrication SEM of all military lots AMCC Confidential and Proprietary DS1596 17 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Table 1. ATM LAN and 100VG AnyLAN Products Products S3011 S3012 S3020 S3021 S3027 S3028 S3029 S2100 Function SONET/ATM/ E-4 Tx SONET/ATM/ E-4 Rx ATM Tx ATM Rx Clock Recovery ATM Transceiver Quad Transceiver 100VG AnyLAN Transceiver Operating Speed 139/155 Mbit/s 139/155 Mbit/s 622 Mbit/s 622 Mbit/s 155/622 Mbit/s 155/622 Mbit/s 155 Mbit/s 120 Mbit/s Data Path 8:1 bit 1:8 bit 8:1 bit 1:8 bit 1 bit 1:8/8:1 bit 1 bit 4:1/1:4 bit Package 80 TEP 80 TEP 52 TEP 52 TEP 20 TSSOP 64 PQFP 64 TQFP 52 PQFP Power Supply +5V +5V +5V +5V +5V +5V +3.3V +5.0V Table 2. Fibre Channel/Gigabit Ethernet Products Products S2036 S2042 S2043 S2044 S2045 S2046 S2047 S2052 S2053 S2054 S2057 S2058 Function Open Fiber Control 10-bit Compliant Fibre Channel 10-bit Compliant Fibre Channel GLM Compliant Fibre Channel Transmitter GLM Compliant Fibre Channel Receiver Gigabit Ethernet Transmitter Gigabit Ethernet Receiver 10-bit Compliant Fibre Channel Gigabit Ethernet and Fibre Channel Transceiver Gigabit Ethernet and Fibre Channel Transceiver Port Bypass Circuit Port Bypass and Repeater Operating Speed 266/531/1062 Mbit/s 266/531/1062 Mbit/s 266/531/1062 Mbit/s 266/531/1062 Mbit/s 266/531/1062 Mbit/s 1250 Mbit/s 1250 Mbit/s 1062/1250 Mbit/s 1062/1250 Mbit/s 1062/1250 Mbit/s Up to 1500 Mbit/s 1062 Mbit/s Data Path N/A 10:1/20:1 bit 1:10/1:20 bit 10:1/20:1 bit 1:10/1:20 bit 10:1/20:1 bit 1:10/1:20 bit 10:1/1:10 bit 10:1/1:10 bit 10:1/1:10 bit 1:1 bit 1:1 bit Package 28 SOIC 52 PQFP 52 PQFP 52 PQFP 52 PQFP 52 PQFP 52 PQFP 64 PQFP 64 PQFP 64 PQFP 20 TSSOP 28 SOIC Power Supply +5V +5V/+3.3V +3.3V +3.3V +3.3V +3.3V +3.3V +3.3V +3.3V +3.3V +3.3V +3.3V See Network Products data book or http://www.amcc.com. AMCC Confidential and Proprietary DS1596 18 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Table 3. HIPPI Products Products S2020 S2021 Function HIPPI Source HIPPI Destination Operating Speed 800 Mbit/s 800 Mbit/s Data Path 32 bit 32 bit Package 225 PGA/208 TEP 225 PGA/208 TEP Power Supply -5.2/+5V -5.2/+5V See Network Products data book or http://www.amcc.com. Table 4. PCI Products Products S5920 S5920DK1 S5933 S5933DK1 S59CD-ROM1 Function Target PCI Interface Developer's Kit Master/Slave PCI Controller Developer's Kit Software CD-ROM Description General Purpose PCI Bus Interface Chip Software/Hardware ISA Bus, SRAM,PLD and User Developer Kit General Purpose PCI Bus Controller Software/Hardware PLD and User Prototype Developer Kit S5920/S5933 Developer Kit Software Code Examples, Drivers and Utility Programs Table 5. SONET/SDH/ATM Products Products S3005 S3006 S3014 S3015 S3016 S3017 S3018 S3019 S3025 S3026/27 S3028 S3029 S3030/31 S3032 S3033 S3035 Function SONET/ATM/ E-4 Tx SONET/E-4 Rx Clock Recovery E4/OC-3/STM-1 Tx E4/OC-3/STM-1 Rx SONET/SDH Tx SONET/SDH Rx SONET/SDH XCVR w/CDR Clock Recovery Clock Recovery SONET/SDH Transceiver Quad Transceiver E4/OC-3/STM-1 ATM XCVR SONET/SDH/ATM XCVR w/ CDR SONET/SDH/ATM XCVR SONET/SDH/ATM XCVR w/ Dual I/O Operating Speed 139/155/622 Mbit/s 139/155/622 Mbit/s 155/622 Mbit/s 139/155 Mbit/s 139/155 Mbit/s 622 Mbit/s 622 Mbit/s 155/622 Mbit/s 622 Mbit/s 155/622 Mbit/s 155/622 Mbit/s 155 Mbit/s 139/155 Mbit/s 155/622 Mbit/s 155/622 Mbit/s 155/622 Mbit/s Data Path 8:1 bit 1:8 bit 1 bit 1 bit 1 bit 8:1 bit 1:8 bit 8:1/1:8 bit 1 bit 1 bit 8:1/1:8 bit 1 bit 1 bit; 4:1/1:4 bit 8:1/1:8 bit 8:1/1:8 bit 8:1/1:8 bit Package 68 LDCC 80 TEP 68 LDCC 80 TEP 44 PLCC 52 TEP 52 TEP 52 TEP 52 TEP 80 PQFP 20 TSSOP 20 TSSOP 64 PQFP 64 TQFP 100 PQFP 64 PQFP 64 TQFP 80 PQFP Power Supply -4.5/5.0V -4.5/5.0V -5.2/+5.0V +5V +5V +5V +5V +3.3V +5V +5V +5V +3.3V +5V +3.3V +3.3V +3.3V See Network Products data book or http://www.amcc.com. AMCC Confidential and Proprietary DS1596 19 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Table 6. SONET/SDH/ATM Products (continued) Products S3040 S3041 S3042 S3043 S3044 S3045 S3047 Function SONET/SDH Clock Recovery Unit SONET/SDH Transmitter SONET/SDH Demux SONET/SDH Transmitter SONET/SDH Demux SONET/SDH OC-12 to OC-48 SONET/SDH Clock and Data Recovery w/ no Ref. Clock Operating Speed 2.5 Gbit/s 2.5 Gbit/s 2.5 Gbit/s 2.5 Gbit/s 2.5 Gbit/s 77/311 Mbit/s 2.5 Gbit/s Data Path 1 bit 8:1 bit 1:8 bit 16:1 bit 1:16 bit 32:8/8:32bit 1 bit Package 32 TQFP 100 TQFP 100 TQFP 80 PQFP 80 PQFP 208 PQFP 32 TQFP Power Supply +5.0V +3.3V +3.3V +3.3V +3.3V +3.3V +5.0V See Network Products data book or http://www.amcc.com. Table 7. Crosspoint Switch Products Products S2016 S2024 S2025 S2028 Function Crosspoint Switch Crosspoint Switch Crosspoint Switch Crosspoint Switch Operating Speed 1.5 Gbit/s 600/800 Mbit/s 1.5 Gbit/s 1.5 Gbit/s Data Path 16 X16 32 X 32 32 X 32 33 X 32 Package 120 TEP 196 LDCC 196 LDCC 224 LDCC Power Supply +5V -5.2/+5V +5V +3.3V See Network Products data book or http://www.amcc.com. Table 8. Precision Clocking Products Output Frequency with Respect to Input Frequency Number of Outputs Divide-by-1 10 10 20 10 6 20 Number of Outputs Divide-by-2 10 10 N/A N/A 8 N/A Power Supply +5V +5V +5V +5V +5V +3.3V Output Levels TTL LVTTL LVTTL LVTTL LVTTL LVTTL P/N SC3506 SC3306 SC3308 SC3318 SC3368 S3LV308 Total Outputs 20 20 20 10 14 20 Package 52 PQFP 52 PQFP 52 PQFP 28 SOIC 28 SOIC 52 PQFP See Network Products data book or http://www.amcc.com. AMCC Confidential and Proprietary DS1596 20 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Table 9. Clock Generator and Synthesizer Products Output Frequency with Respect to Input Frequency Number of Selectable Output Relationships 21 21 21 7x4 Banks of 3 Outputs Multiply 2-32 Divide 2-16 1 P/N S4402 4403 4405 S4406 Description Multiphase Clock Generator Multiphase Clock Generator Multiphase Clock Generator Clock Generator with delay Adj.& Invert Clock Synthesizer RAMBUS-TM Compatible Clock Generator Input Reference TTL TTL PECL/TTL TTL Number 6 10 61 12 Type TTL TTL TTL PECL TTL Maximum Frequency 80 80 80 160 66 Minimum Delay Adjust Increment 3.2ns 3.2ns 3.2ns 4 ns @ 66 MHz 4503 S4506 S4507 XTAL XTAL 21 2 TTL PECL Rambus Compatible 80 300 250 300 N/A N/A See Network Products data book or http://www.amcc.com. Table 10. ASIC Standard Cell Products Family Name Micro-power Technology 1 Micron Bipolar Operating Speed Up to 2.5GHz Equivalent Gates Up to 4000 Number of I/O Up to 200 Analog Functions • 2.5 GHz PLL • 1 GHz Timing Vernier • Custom Analog Power Supplies Options • • • • +5V +5V/-5V -5V +3.3V See Network Products data book or http://www.amcc.com. AMCC Confidential and Proprietary DS1596 21 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Table 11. ASIC Logic Array Products Part Number Q20004 Q20010 Q20025 Q20045 Q20080 Q20120 Q20P010 Q20P025 Q20M100 Technology 1 Micron Bipolar 1 Micron Bipolar 1 Micron Bipolar 1 Micron Bipolar 1 Micron Bipolar 1 Micron Bipolar 1 Micron Bipolar 1 Micron Bipolar 1 Micron Bipolar Equivalent Gates (Full Adder Method) 671 1469 4032 6782 11242 18777 973 3272 13475 Number of I/O 30 68 102 130 164 200 34 51 195 Structured Arrayed Blocks None None None None None None 1.25 GHz PLL 1.25 GHz PLL RAM See Network Products data book or http://www.amcc.com. AMCC Confidential and Proprietary DS1596 22 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book ARCHITECTURAL OVERVIEW S5920 Since the S5920 is a PCI Target or Slave device only, its cost is significantly less than PCI Bus Master solutions. The S5920 is PCI purposed 2.2 compliant and can support data transfer rates up to 132 Mbytes/sec. Burst transfers and single data transfers are both supported. Figure 1 shows the block diagram for the S5920. Many additional S5920 features offer the user easier hardware and software implementation. Up to four memory or I/O size definable blocks, referred to as Pass-Thru‘ regions, are provided for multiple device configurations. Data transfers via a Pass-Thru region can be performed either direct to the Add-On bus or through two 32-Byte burstable FIFOs. Added read prefetch and programmable FIFO wait state features allow the user to tune system performance. The PassThru data channel also supports an active/passive mode bus interface. Passive mode requires the designer to transfer data by externally driving the AddOn bus. Active mode minimizes design components by enabling internal logic to drive or acquire the AddOn bus to read or write data independently. Active mode provides programmable wait state generation for slower Add-On designs. Two 32-bit mailbox registers are implemented for additional data or user-defined status/command transfers. Each mailbox may be examined for empty or full, at the byte level, through a mailbox status register. Mailbox transfers can be either register style or hardware direct. Dedicated external mailbox data and strobe pins are provided for direct hardware read/writes and allow Add-On to PCI interrupt capabilities. A direct Add-On to a PCI bus interrupt pin is incorporated, adding design flexibility. The S5920 supports a two-wire serial nvRAM. This allows the designer to customize the device configuration to be loaded during power-up initialization. An expansion BIOS may also be contained in the nvRAM. and Add-On Operation Registers. All of these registers are user configurable through their associated buses and from the external nvRAM. The following sections provide a brief overview of each register group and the nvRAM interface. PCI Configuration Registers All PCI compliant devices are required to provide a group of PCI configuration registers. These registers are polled by the host BIOS system during power-up initialization. They contain specific device and product information such as Vendor ID, Device ID, Subsystem Vendor ID, memory requirements, etc. These registers are located in the S5920 and are either initialized with predefined default values or user customized definitions contained in the external nvRAM. PCI Bus Accessible Registers The second group of registers are the PCI Operation Registers. This group of registers is accessible to the PCI Bus. These are the primary registers through which the PCI Host configures the S5920 operation and communicates with the Add-On Bus. These registers encompass the PCI bus mailboxes, Pass-Thru/ FIFO data channel and Status/ Control registers. Add-On Bus Accessible Registers The last register group consists of the Add-On Operation Registers. This group of registers is accessible via the Add-On Bus. These are the primary registers through which the Add-On application configures S5920 operation and communicates with the PCI Bus. These registers encompass the Add-On bus mail boxes, Pass-Thru/FIFO Registers and Status/Control Registers. SERIAL NON-VOLATILE INTERFACE Previously indicated, the S5920 contains the required set of PCI Configuration Registers. These registers can be initialized with default values or with customized values contained in an external nvRAM. The nvRAM allows Add-On card manufacturers to initialize the S5920 with their specific Vendor ID values, along with other desired S5920 operation characteristics. S5920 REGISTER ARCHITECTURE S5920 communications, control and configuration is performed through three primary groups of registers: PCI Configuration Registers, PCI Operation Registers AMCC Confidential and Proprietary DS1596 23 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Figure 3. S5920 Pinout S5920 PCLK INTA# RST# AD[31:0] C/BE[3:0]# BPCLK ADCLK SYSRST# IRQ# ADDINT# DQ[31:0] SELECT# ADR[6:1] BE[3:0]# RD# WR# PTATN# PTBURST# PTNUM[1:0] PTBE[3:0]# PTADR# PTWR PTRDY#/WAIT# DXFER# PTMODE DQMODE MD[7:0] LOAD# MDMODE SDA SCL Add-On Bus Timing/Interrupts Add-On Data Bus PCI 2.1 Local Bus FRAME# DEVSEL# IRDY# TRDY# IDSEL# STOP# LOCK# PAR PERR# SERR# S5920 Data Access Control Pass-Thru Control/ Access S5920 Control FLT# Add-On Bus Control Mail Box Access/Control Serial Bus Config/BIOS Opt. AMCC Confidential and Proprietary DS1596 24 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book MAILBOX OPERATION The mailbox registers are divided into two 4-byte sets. Each set is dedicated to one bus for data transfer to the other bus. Figure 3 shows a block diagram of the mailbox section of the S5920. The provision of mailbox registers provides data or user defined command/status transfer capability between two buses. An empty/ full indication for each mailbox register, at the byte level, is determined by polling a status register accessible to both the PCI and Add-On buses. Providing mailbox byte level full indications allows greater flexibility in 8-, 16-or 32-bit designs; i.e., transferring a single byte in 8-bit Add-On bus without requiring the assembly or disassembly of 32-bit data. A mailbox byte level interrupt feature for PCI or AddOn buses is provided. Bit locations configured within the S5920 operation registers can select which mailbox byte is to generate an interrupt when the mailbox is written to. Interrupts can be generated to the PCI or Add-On buses. PCI bus interrupts may also be generated from direct hardware interfacing due to a unique S5920 feature. The Add-On mailbox is hardware accessible via a set of dedicated device pins. A single load pulse latches data into the mailbox generating an interrupt, if enabled. PASS-THRU OPERATION Pass-Thru region accesses can execute PCI bus cycles in real time or through an internal FIFO. Real time operation allows the PCI bus to directly read or write to Add-On bus resources. The S5920 allows the designer to declare up to four individual Pass-Thru regions. Each region may be defined as 8, 16 or 32 bits wide, mapped into memory or I/O system space and may be up to 512 MB in size. Figure 4 shows a b as i c bl o c k d i a gr am o f t h e S 5 9 20 P a s s - Th r u architecture. Host communications to the Pass-Thru data channel utilizes dedicated Add-On bus pins to signal that a PCI read or write has been requested. User logic decodes these signals to determine if it must read or write data to the S5920 to satisfy the PCI request. Information decoded includes: PCI read/write transaction request, the byte lanes involved, the specific Pass-Thru region accessed and whether the request is a burst or single cycle access. Pass-Thru operation supports single PCI data cycles and PCI data bursts. During PCI burst operations, the S5920 is capable of transferring data at the full PCI bandwidth. Should slower Add-On logic be imple mented, the S5920 will issue a PCI bus retry until the requested transfer is completed. Figure 4. Mailbox Block Diagram 88 8 8 Mailbox Mailbox Mailbox Mailbox Byte 0 Byte 1 Byte 2 Byte 3 32 PCI Bus PCI Decode Control Mailbox Status Register 32 Add-On Decode Control Mailbox Mailbox Mailbox Mailbox Byte 0 Byte 1 Byte 2 Byte 3 88 8 8 8/16/32-Bit Add-On Bus DS1596 AMCC Confidential and Proprietary 25 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book To increase data throughput, the Pass-Thru channel incorporates two 32-byte FIFOs. One FIFO is dedi cated to PCI read data while the other is dedicated to PCI write data. Enabling the write FIFO allows the S5920 to accept zero wait state bursts from the PCI bus regardless of the Add-On bus application design speed. Figure 4 illustrates the Pass-Thru block. Enabling the read FIFO allows data to be optionally prefetched from the Add-On Bus. This can greatly improve performance of slow Add-On bus designs. PCI read cycles can be performed with zero wait Figure 5. Pass-Thru Block Diagram states since data has been prefetched into the FIFO. Either of the write/read FIFOs can be disabled or enabled to tune system performance. The Add-On bus can be operated in two different modes: active or passive. The passive mode of operation mimics that of the S5933 Add-On bus operation. The user design drives S5920 pins to read or write data. In active mode, the Add-On bus is driven from an S5920 internal state machine. This reduces component count in cost-sensitive designs. Active mode also incorporates programmable wait states from 0 to 7. 32 Pass-Thru Register Endian Conv. 32-Byte FIFO 32 PCI Bus PCI Decode Control Add-On Decode Control Status/CTRL Register 32 32-Byte FIFO Endian Conv. Pass-Thru Register 32 8/16/32-Bit Add-On Bus DS1596 AMCC Confidential and Proprietary 26 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Signal Type Definitions The following signal types are taken from the PCI Bus Specification. in out t/s s/t/s Input is a standard input-only signal. Totem Pole Output is a standard active driver. Tri-State ®is a bi-directional, tri-state input/output pin. Sustained Tri-State is an active low tri-state signal owned and driven by one and only one agent at a time. The agent that drives an s/t/s pin low must drive it high for at least one clock before letting it float. A new agent cannot start driving an s/t/s signal any sooner than one clock after the previous owner tri-states it. A pull-up is required to sustain the inactive state until another agent drives it, and must be provided by the central source. Open Drain allows multiple devices to share as a wire-OR. o/d Each signal that assumes the logic low state when asserted is followed by the pound sign (#). Example: TRDY# signal is asserted low when the target is ready to complete a data transfer. Signals that are not followed by the pound sign are asserted when they assume the logic high state. The following designations are used throughout this book when referring to the size of data objects: A BYTE is an 8-bit object. A WORD is a 16-bit, or 2-byte object. A DWORD is a double word and is a 32-bit or 4-byte object. All hex numbers are followed by an “h”. Examples: 9A4Fh 0110h All binary numbers are followed by an “b”. Examples: 1010b 0110b All decimal numbers are followed by an “d”. Examples: 4356d 1101d Note: Tri-State® is a Registered Trademark of National Semiconductor. AMCC Confidential and Proprietary DS1596 27 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Figure 6. S5920 Pin Assignment DQ0 56 55 54 52 48 47 46 44 42 40 39 38 36 35 34 32 14 12 8 7 6 4 3 2 158 156 155 154 152 148 147 146 142 139 58 43 28 15 159 16 DQ1 AD0 AD1 AD2 AD3 AD4 AD5 AD6 AD7 AD8 AD9 AD10 AD11 AD12 AD13 AD14 AD15 AD16 AD17 AD18 AD19 AD20 AD21 AD22 AD23 AD24 AD25 AD26 AD27 AD28 AD29 AD30 AD31 CLK RST# INTA# C/BE0# C/BE2# C/BE3# FRAME# DEVSEL# IRDY# TRDY# IDSEL STOP# LOCK# SELECT# 27 24 26 PAR PERR# SERR# PTNUM0 PTNUM1 59 138 135 136 113 149 29 DQMODE FLT# RSVD3 RSVD4 RSVD2 RSVD5 RSVD1 PTATN# PTBURST# PTADR# PTWR 1 9 10 17 21 30 41 49 50 70 90 106 110 121 130 137 141 150 153 GND GND GND GND GND GND GND GND GND GND GND GND GND GND GND GND GND GND GND SDA SCL 11 31 33 51 71 91 103 111 129 131 151 VCC VCC VCC VCC VCC VCC VCC VCC VCC VCC VCC MD0 MD1 MD2 MD3 MD4 MD5 MD6 MD7 LOAD# MDMODE 57 61 69 73 81 89 97 101 109 143 127 128 PTRDY#/WAIT# PTMODE 114 112 107 108 115 104 PTBE0# PTBE1# PTBE2# PTBE3# 122 123 116 118 119 120 WR# RD# 75 74 72 BE0# BE1# BE#2 22 23 BE3#/ADR1 87 63 62 60 68 67 66 64 132 BPCLK ADCLK IRQ# ADDINT# SYSRST# DXFR# C/BE1# ADR2 ADR3 ADR4 ADR5 ADR6 140 134 124 102 126 144 DQ2 DQ3 DQ4 DQ5 DQ6 DQ7 DQ8 DQ9 DQ10 DQ11 DQ12 DQ13 DQ14 DQ15 DQ16 DQ17 DQ18 DQ19 DQ20 DQ21 DQ22 DQ23 DQ24 DQ25 DQ26 DQ27 DQ28 DQ29 DQ30 DQ31 100 99 98 96 95 94 92 88 86 84 83 82 80 79 78 76 157 145 133 125 117 105 93 85 77 65 53 45 37 25 13 5 Data Bus Bus Controls PCI Bus 20 18 19 160 Register Access Controls S5920 Controls Pass-Thru Data Controls Mail Box Bus Power & Ground NVRAM Bus Add-On User Bus AMCC Confidential and Proprietary DS1596 28 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book PCI BUS SIGNALS The following sets of signals represent the interface pins available for the S5920 to PCI bus. Table 12. PCI Bus Address and Data Signal Signal AD[31:0] Type t/s Description Address/Data. Address and data are multiplexed on the same PCI bus pins. A PCI bus transaction consists of an address phase followed by one or more data phases. An address phase occurs on the PCLK cycle in which FRAME# is asserted. A data phase occurs on the PCLK cycles in which IRDY# and TRDY# are both asserted. Command/Byte Enable. Bus commands and byte enables are multiplexed on the same pins. These pins define the current bus command during an address phase. During a data phase, these pins are used as Byte Enables, with C/BE[0]# enabling byte 0 (LSB) and C/BE[3]# enabling byte 3 (MSB). C/BE[3:0] 0000 0001 0010 0011 0100 0101 0110 0111 1000 1001 1010 1011 1100 1101 1110 1111 PAR t/s Command Type Interrupt Acknowledge Special Cycle I/O Read I/O Write Reserved Reserved Memory Read Memory Write Reserved Reserved Configuration Read Configuration Write Memory Read Multiple Dual Address Cycle Memory Read Line Memory Write and Invalidate C/BE[3:0]# in Parity. Parity is always driven as even from all AD[31:0] and C/BE[3:0]# signals. The parity is valid during the clock following the address phase and is driven by the bus master. During a data phase for write transactions, the bus master sources this signal on the clock following IRDY# active; during a data phase for read transactions, this signal is driven by the target and is valid on the clock following TRDY# active. The PAR signal has the same timing as AD[31:0], delayed by one clock. AMCC Confidential and Proprietary DS1596 29 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Table 13. PCI Bus System Signals Signal PCLK RST# Type in in Description PCI Clock. The rising edge of this signal is the reference upon which all other signals are based except for RST# and INTA#. The maximum PCLK frequency for the S5920 is 33 MHz and the minimum is DC (0 Hz). Reset brings the S5920 to a known state: - All PCI bus output signals tri-stated. - All open drain signals (i.e., SERR#) floated. - All registers set to their factory defaults. - Pass-Thru is returned to an idle state. - All FIFOs emptied. Table 14. PCI Bus Data Transfer Control Signals Signal FRAME# Type in Description Frame. This signal is driven by the current bus master to indicate the beginning and duration of a bus transaction. When FRAME# is first asserted, it indicates a bus transaction is beginning with a valid addresses and bus command present on AD[31:0] and C/BE[3:0]. Data transfers continue while FRAME# is asserted. FRAME# de-assertion indicates the transaction is in a final data phase or has completed. Initiator Ready. This signal is always driven by the bus master to indicate its ability to complete the current data phase. During write transactions, it indicates AD[31:0] contains valid data. Target Ready. This signal is driven by the selected target to indicate the target is able to complete the current data phase. During read transactions, it indicates AD[31:0] contains valid data. Wait states occur until both TRDY# and IRDY# are asserted together. Stop. The Stop signal is driven by a selected target and conveys a request to the bus master to stop the current transaction. Lock. The lock signal provides for the exclusive use of a resource. The S5920 may be locked by one master at a time. Initialization Device Select. This pin is used as a chip select during configuration read or write transactions. Device Select. This signal is driven by a target decoding and recognizing its bus address. This signal informs a bus master whether an agent has decoded a current bus cycle. Interrupt A. This signal is defined as optional and level sensitive. Driving it low will interrupt to the host. The INTA# interrupt is to be used for any single function device requiring an interrupt capability. IRDY# TRDY# in s/t/s STOP# LOCK# IDSEL DEVSEL# INTA# s/t/s in in s/t/s o/d Table 15. PCI Bus Error Reporting Signals Signal PERR# Type s/t/s Description Parity Error. Only for reporting data parity errors for all bus transactions except for Special Cycles. It is driven by the agent receiving data two clock cycles after the parity was detected as an error. This signal is driven inactive (high) for one clock cycle prior to returning to the tri-state condition. System Error. Used to report address and data parity errors on Special Cycle commands and any other error condition having a catastrophic system impact. Special Cycle commands are not supported by the S5920. SERR# o/d AMCC Confidential and Proprietary DS1596 30 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book ADD-ON BUS AND S5920 CONTROL SIGNALS The following sets of signals represent the interface signals available for the user Add-On bus and S5920 control. Table 16. Serial nvRAM Interface Signals Signal SCL Type o/d-out Description Serial Clock. This clock provides timing for all transactions on the two-wire serial bus. The S5920 drives this signal when performing as a serial bus master. SCL operates at the maximum allowable clock speed and enters the high Z state when FLT# is asserted or the serial bus is inactive. Serial Data/Address. This bi-directional signal carries serial address and data information between nvRAMs and the S5920. This pin enters high Z state when FLT# is asserted or the serial bus is inactive. Reserved. Must be left open. SDA Pin 135 o/d in Table 17. Direct Mailbox Access Signals Signal MDMODE Type in Description Mailbox Data Mode. The MD[7:0] signal pins are always inputs when this signal is high. The MD[7:0] signal pins are defined as inputs and outputs under LOAD# control when MDMODE is low. This pin is provided for software compatibility with the S5933. New designs should permanently connect this signal low. This signal is connected to an internal pull-up. MD[7:0] is defined as an input bus when this signal is low. The next rising edge of the ADCLK will latch MD[7:0] data into byte three of the Add-On outgoing mailbox. When LOAD# is high and MDMODE is low, MD[7:0] are defined as outputs displaying byte three of the PCI outgoing mailbox. This signal is connected to an internal pull-up. Mailbox Data bus. The mailbox data registers can be directly accessed using the LOAD# and MDMODE signals. When configured as an input, data byte three of the PCI incoming mailbox is directly written to from these pins. When configured as an output, data byte three of the PCI outgoing mailbox is output to these pins. All MD[7:0] signals have an internal pull-up. LOAD# in MD[7:0] t/s AMCC Confidential and Proprietary DS1596 31 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book USER ADD-ON BUS PIN DESCRIPTIONS The following sets of signals represent the interface pins available for the Add-On bus. The following defines three signal groups: S5920 register access signals, Pass-Thru channel signals, and general Add-On bus signals. Table 18. Pass-Thru Data Channel Pins Signal PTMODE Type in Description Pass-Thru Mode. Configures the Pass-Thru data channel operation. High configures the S5920 in Passive mode allowing other devices to read/write data bus data. Low configures the S5920 in Active mode. This mode allows the S5920 to actively drive signals and data onto the data bus. This signal is connected to an internal pull-up. Pass-Thru Attention. Signals a decoded PCI to Pass-Thru region bus cycle. PTATN# is generated to signal that Add-On logic Pass-Thru data must be read from or written to the S5920. Pass-Thru Burst. Informs the Add-On bus that the current Pass-Thru region decoded PCI bus cycle is a burst access. Pass-Thru Ready/Pass-Thru Wait. During passive mode, the signal is referred to as PTRDY# and is asserted low to indicate Add-On logic has read/written data in response to a PTATN# signal. During active mode operation, the signal is referred to as WAIT# and can be driven low to insert wait states or hold the S5920 from clocking data onto the data bus. PTRDY# or WAIT# is synchronous to ADCLK. Pass-Thru Number. Identifies which of the four Pass-Thru regions the PTATN# read/write is requesting. Only valid for the duration of PTATN# active. 00 = Base Address Register 1, 01 = Base Address Register 2, 10 = Base Address Register 3, 11 = Base Address Register 4. Pass-Thru Byte Enables. During a PCI to Pass-Thru read, PTBE[3:0] indicate which bytes of a DWORD are to be written into. During a PCI to Pass-Thru write, these pins indicate which bytes of a DWORD are valid to read. PTBE[3:0]# are only valid while PTATN# is asserted. Pass-Thru Address. Is an input when in passive mode. When asserted, the 32-bit Pass-Thru address register contents are driven onto the DQ[31:0] bus. All other Add-On control signals must be inactive during the assertion of PTADR# in passive mode. In active mode, becomes an output and indicates a Pass-Thru address is on the DQ bus. The DQMODE signal does not affect DQ bus width while the Pass-Thru address is driven. Pass-Thru Write. This signal indicates whether the current PCI to Pass-Thru bus transaction is a read or write cycle. Valid only when PTATN# is active. ACTIVE Transfer complete. When in ACTIVE mode, this output is asserted at the end of every 816- or 32-bit data transfer cycle. This signal is not used in Passive mode. PTATN# PTBURST# PTRDY#/ WAIT# out out in PTNUM[1:0] out PTBE[3:0]# out PTADR# t/s PTWR DXFER# out out AMCC Confidential and Proprietary DS1596 32 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Table 19. S5920 Add-On Bus Register Access Pins Signal DQ[31:0] ADR[6:2] Type t/s in Description Address/Data bus. The 32-bit Add-On data bus. The DQMODE signal configures the bus width for either 32 or 16 bits. All DQ[31:0] signals have an internal pull-up. Address [6:2]. These inputs select which S5920 register is to be read from or written to. To be used in conjunction with SELECT#, BE[3:0]# and WR# or RD#. The register addresses are as follows: ADR[6:2] 00011 00111 01010 01011 01101 01110 01111 10000 Register Name Add-On Incoming Mailbox Register Add-On Outgoing Mailbox Register Add-On Pass-Thru Address Register Add-On Pass-Thru Data Register Add-On Mailbox Status Register Add-On Interrupt Control Register Add-On Reset Control Register Pass-Thru/FIFO Configuration Register Note: ADR[6:2] bits begin at bit position two. All references to an address, in hex, adds bits 0 and 1 as zeros. Example: The Add-On incoming mailbox register is referenced as 0Ch. BE[2:0]# in Byte Enable 2 through 0. Provides individual read/write byte enabling during register read or write transactions. BE2# enables activity over DQ[23:16], BE1# enables DQ[15:8], and BE0# enables DQ[7:0]. During read transactions, these pins enable the output driver for each byte lane; for write transactions, they serve as an input enable to perform the write to each byte lane. Byte Enable [3] for a 32-bit bus width / Address [1] for a 16-bit bus width. BE3#, enables DQ[31:24] input drivers for writing data to registers identified by ADR[6:2] and enables DQ[31:24] output drivers to read registers identified by ADR[6:2]. To be used in conjunction with SELECT# and RD# or WR#. ADR1, selects the upper or lower WORD of a DWORD when a 16-bit-wide bus is selected. 1 = upper, 0 = lower. Select. Enables internal S5920 logic to decode WR#, RD# and ADR[6:2] when reading or writing to any Add-On register. Write Enable. Asserting this signal writes DQ bus data byte(s) selected by BE[3:0]# into the S5920 register defined by SELECT# and ADR[6:2]. Read Enable. Asserting this signal drives data byte(s) selected by BE[3:0]# from the S5920 register defined by SELECT# and ADR[6:2] onto the DQ bus. DQ Mode. Defines the DQ bus width when accessing data using WR#, RD#, SELECT# and ADR[6:2]#. Low = 32-bit wide DQ bus. High = 16-bit wide DQ bus. When high, the signal BE3# is reassigned to the ADR1 signal and only DQ[15:0] is active. Note: This pin only affects DQ Bus Width for S5920 Data Registers. This pin has no effect on accesses DQ Bus Width. For the Pass-Thru data register (APTD, ADR = 2Ch). The width of the DQ bus is determined by the region-size bits in the corresponding Base Address Register. In addition, DQMODE has no effect when using the direct-access pin PTADR#. When PTADR# is asserted, all 32 bits of the Pass-Thru address are provided. BE3# / ADR1 in SELECT# WR# RD# DQMODE in in in in AMCC Confidential and Proprietary DS1596 33 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Table 20. Add-On Bus General Pins Signal SYSRST# BPCLK ADCLK IRQ# ADDINT# Type out out in out in Description System Reset. An active-low buffered PCI bus RST# output signal. The signal is asynchronous and can be asserted through software from the PCI host interface. Buffered PCI Clock. This output is a buffered form of the PCI bus clock and has all of the behavioral characteristics of the PCI clock (i.e., DC-to-33 MHz capability). Add-On Clock. All internal S5920 Add-On bus logic is synchronous to this clock. The clock is asynchronous to the PCI bus logic unless connected to the BPCLK signal. Interrupt Request. This output signals to Add-On logic that a significant event has occurred as a result of activity within the S5920. Add-On interrupt. When enabled and asserted, this input will cause a PCI bus interrupt by driving INTA# low. The input is level sensitive and can be driven by multiple sources. This signal is connected to an internal pull-up. Float. Floats all S5920 output signals when asserted. This signal is connected to an internal pull-up For factory use only. Must be left open. For factory use only. Must be left open. For factory use only. Must be left open. For factory use only. Must be left open. For factory use only. Must be left open. FLT# Pin 149 Pin 136 Pin 135 Pin 113 Pin 29 in X X X X X AMCC Confidential and Proprietary DS1596 34 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Each PCI bus device contains a unique 256-byte region called its configuration header space. Portions of this configuration header are mandatory in order for a PCI agent to be in full compliance with the PCI specification. This section describes each of the configuration space fields—its address, default values, initialization options, and bit definitions—and also provides an explanation of its intended usage. Table 21. Configuration Registers Address Offset 00h–01h 02h–03h 04h–05h 06h–07h 8h0 09h–0Bh 0Ch 0Dh 0Eh 0Fh 10h–27h 28h–2Bh 2Ch–2Dh 2Eh–2Fh 30h–33h 34h–3Bh 3Ch 3Dh 3Eh 3Fh 40h–FFh 1. BADR 5 is not implemented in the S5920. Abbreviation VID DID PCICMD PCISTS RID CLCD CALN LAT HDR BIST BADR0-BADR5 – SVID SID XROM – INTLN INTPIN MINGNT MAXLAT – Register Name Vendor Identification Register Device Identification Register Command Register Status Register Revision Identification Register Class Code Register Cache Line Size Register Latency Timer Register Header Type Register Built-in Self-test Register Base Address Registers (0-5)1 Reserved Subsystem Vendor Identification Register Subsystem Identification Register Expansion ROM Base Address Register Reserved Interrupt Line Register Interrupt Pin Register Minimum Grant Register Maximum Latency Register Not used AMCC Confidential and Proprietary DS1596 35 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book PCI Configuration Space Header 31 24 23 DEVICE ID STATUS CLASS CODE BIST HEADER TYPE LATENCY TIMER 16 15 87 VENDOR ID COMMAND REV ID CACHE LINE SIZE 00 00 04 08 0C 10 14 18 1C 20 24 28 SUBSYSTEM VENDOR ID 2C 30 34 38 INTERRUPT LINE 3C BASE ADDRESS REGISTER #0 BASE ADDRESS REGISTER #1 BASE ADDRESS REGISTER #2 BASE ADDRESS REGISTER #3 BASE ADDRESS REGISTER #4 BASE ADDRESS REGISTER #5 RESERVED = 0's SUBSYSTEM ID EXPANSION ROM BASE ADDRESS RESERVED = 0's RESERVED = 0's MAX_LAT MIN_GNT INTERRUPT PIN LEGEND EPROM IS DATA SOURCE (READ ONLY) CONTROL FUNCTION EPROM INITIALIZED RAM (CAN BE ALTERED FROM PCI PORT) EPROM INITIALIZED RAM (CAN BE ALTERED FROM ADD-ON PORT) HARD-WIRED TO ZEROES Note: Some registers are a combination of the above. See individual sections for full description. AMCC Confidential and Proprietary DS1596 36 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book VENDOR IDENTIFICATION REGISTER (VID) Register Name: Address Offset: Power-up value: Boot-load: Attribute: Size: Vendor Identification 00h-01h 10E8h (AMCC’s) External nvRAM offset 040h-41h Read Only 16 Bits This register is reserved for the manufacturer’s device identification number (VID). VID numbers are assigned to each PCI device manufacturer by an international organization known as the PCI Special Interest Group (SIG). This ensures PCI device uniqueness among all manufacturers. This register defaults to AMCC's VID during power-on initialization. The default value can be changed to another valid VID when an external non-volatile device is used for boot loading. Figure 7. Vendor Identification Register 15 10E8h Vendor Identification Register (RO) 0 Bit 15.0 Description Vendor Identification Number: AMCC’s 16-bit value is 10E8h AMCC Confidential and Proprietary DS1596 37 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book DEVICE IDENTIFICATION REGISTER (DID) Register Name: Address Offset: Power-up value: Boot-load: Attribute: Size: Device Identification 02h-03h 5920h External nvRAM offset 042h-43h Read Only 16 bits This register is reserved for the manufacturer’s device identification number (DID). DID numbers are maintained and issued by the registered owner of the VID number programmed into a PCI device. AMCC will issue a DID number to manufacturers using the AMCC VID. This maintains PCI device uniqueness among all manufacturers. to be in compliance. This register defaults to AMCC's DID during power-on initialization. The default value can be changed to another valid DID when an external nonvolatile device is used for boot loading. Figure 8. Device Identification Register 15 5920h Device Identification Register (RO) 0 Bit 15.0 Description Device Identification Number: AMCC’s temporary end user value 5920h AMCC Confidential and Proprietary DS1596 38 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book PCI COMMAND REGISTER (PCICMD) Register Name: Address Offset: Power-up value: Boot-load: Attribute: Size: PCI Command 04h-05h 0000h not used Read/Write (R/W on 4 bits, R/O for all others) 16 bits This 16-bit register provides basic control over a device’s ability to respond to or perform PCI accesses. This register is defined by the PCI specification and its implementation is required of all PCI devices. Four of the ten implemented bits are required by the S5920; those which are not required are hardwired to 0. The definitions for all the fields are provided here for completeness. Figure 9. PCI Command Register 15 Reserved = 00 Fast Back-to-Back SERR# Enable Wait Cycle Enable PERR# Enable Palette Snoop Enable Memory Write and Invalidate Enable Special Cycle Enable Bus Master Enable Memory Access Enable I/O Access Enable 9 0 8 X 7 0 6 X 5 0 4 0 3 0 2 0 1 X 0 X AMCC Confidential and Proprietary DS1596 39 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Bit 15:10 9 8 Reserved. Hardwired to 0. Fast Back-to-Back Enable. This bit enables fast back-to-back capability for bus master transaction. The S5920 is a target-only device and hardwires this bit to a 0. System Error Enable. Setting this bit to a 1 allows the S5920 to drive the SERR# signal. Setting to a 0 will disable the output driver. The assertion of RESET# will set this bit to a 0. The SERR# pin driven active normally signifies a parity error occurred during a PCI address phase. Wait Cycle Enable. Controls whether a device implements address/data stepping. This bit is hardwired to 0 as the S5920 does not uses stepping. Parity Error Enable. This bit allows the S5920 to drive the PERR# and to generate a SERR# signal. A one allows the parity generation and a 0 will disable generation of a parity error indication. This bit is set to 0 when RESET# is asserted. Palette Snoop Enable. Enables VGA compatible devices to perform palette snooping. This bit is hardwired to a 0 as the S5920 is not a PCI-based VGA device. Memory Write and Invalidate Enable. This bit enables bus masters to generate Memory Write and Invalidate PCI bus commands when set to a 1. When set to 0, bus masters generate memory write commands instead. The S5920 is a PCI target only and therefore hardwires this bit to 0. Special Cycle Enable. Setting this bit to one enables devices monitoring of PCI special cycles. The S5920 does not monitor (or generate) special cycles and hardwires this bit to 0. Bus Master Enable. This bit allows a PCI device to function as a Bus Master. The S5920 is a PCI target device only and hardwires his bit to 0. Memory Space Enable. This bit enables S5920 memory region decodes to any of the five defined base address register memory regions and the Expansion ROM Base Address Register. This bit is cleared to 0 when RESET# is asserted. I/O Space Enable. This bit enables S5920 I/O region decodes to any of the five defined base address register I/ O regions. This bit is cleared to 0 when RESET# is asserted. Description 7 6 5 4 3 2 1 0 AMCC Confidential and Proprietary DS1596 40 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book PCI STATUS REGISTER (PCISTS) Register Name: Address Offset: Power-up value: Boot-load: Attribute: Size: PCI Status 06h-07h 0200h not used Read Only Read/Write Clear 16 bits This register contains PCI device status information. This register is defined by the PCI specification and its implementation is required of all PCI devices. Only applicable bits are used by the S5920; those which are not used are hardwired to 0. Status bits within this register are designated as “write one clear,” meaning that in order to clear a given bit, a 1 must be written. All R/ W/C bits written with a 0 are left unchanged. These bits are identified in Figure 4 as (R/WC). Those which are Read Only are shown as (RO). Figure 10. PCI Status Register 15 X 14 X 13 0 12 0 11 0 10 0 9 1 8 0 7 1 6 0 54 0 0 Reserved = 00's Reserved (RO) 66 Mhz Capable UDF Supported Fast Back-to-Back Capable (RO) Data Parity Reported (RO) DEVSEL# Timing Status 00 = Fast 01 = Medium 10 = Slow 11 = Reserved Signaled Target Abort (R/WC) Received Target Abort (RO) Received Master Abort (RO) Signaled System Error (R/WC) Detected Parity Error (R/WC) AMCC Confidential and Proprietary DS1596 41 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Bit Description 15 14 13 12 11 10:9 8 7 6 5 4:0 Detected Parity Error. This bit is set whenever the S5920 detects a parity error. It is set independent of the state of Command Register Bit 6. The bit is cleared by writing a 1. Signaled System Error. This bit is set whenever the S5920 generates the SERR# signal. This bit can be reset by writing a 1. Received Master Abort. Bus master devices set this bit to indicate a bus master transaction has been terminated due to a master abort. The S5920 is a target device and hardwires this to 0. Received Target Abort. This bit is set by a bus master when its transaction is terminated by a target abort from the currently addressed target device. This bit is required for bus masters and is hardwired to 0 in the S5920. Signaled Target Abort. This bit is set the target device whenever it terminates a transaction with a target abort. The S5920 does not issue target aborts and hardwires this bit to 0. Device Select Timing. These bits are read-only and define the DEVSEL# timing for a target device. The S5920 is a medium PCI device. Data Parity Reported. Only implemented by bus mastering devices to notify a parity error has been detected. This is not applicable to the S5920 and is hardwired to 0. Fast Back-to-back Capable. This read-only bit indicates if a target device supports fast back-to-back transactions. The S5920 supports this feature and hardwires the bit to 1. UDF Supported. 1 = device supports user-definable features. 0 = device does not support user- definable features. The S5920 implements definable memory regions and hardwires this bit to 0. 66 MHz Capable. 1 = device is capable of running at 66 MHz. 0 = device is capable of running at 33 MHz. This bit is hardwired to 0. Reserved. Hardwired to zero. AMCC Confidential and Proprietary DS1596 42 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book REVISION IDENTIFICATION REGISTER (RID) Register Name: Address Offset: Power-up value: Boot-load: Attribute: Size: Revision Identification 08h 00h External nvRAM/EPROM offset 048h R/W 8 Bits This register is reserved for AMCC’s S5920 silicon revision identification number. The register defaults to that value after power up. Write operations from the PCI interface have no effect on the register. Userdefined values can be boot-loaded from an optional external non-volatile. AMCC does not recommend changing the register value. The Sub-system Vendor ID and/or Subsystem ID are intended for end user information. Figure 11. Revision Identification Register 7 00h 0 Revision Identification Number (RO) Bit Description 7:0 Revision Identificaiton Number: Initialized to the S5920 silicon revision. AMCC Confidential and Proprietary DS1596 43 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book CLASS CODE REGISTER (CLCD) Register Name: Address Offset: Power-up value: Boot-load: Attribute: Size: Class Code 09h-0Bh FF0000h External nvRAM offset 049h-4Bh Read Only 24 Bits This 24-bit, read-only register is divided into three onebyte fields: the base class byte at location 0Bh, the sub-class byte at 0Ah, and the programming interface byte at 09h. The default setting for the base class is FFh, which indicates that the device does not fit into the thirteen base classes defined in the PCI Bus Specification. It is possible, however, through use of the external non-volatile memory to change the value of this register. Refer to the PCI specification for details. Figure 12. Class Code Register @0Bh 7 Base Class 0 7 @0Ah 07 Sub-Class @09h 0 Prog I/F AMCC Confidential and Proprietary DS1596 44 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Table 22. Defined Base Class Codes Base-Class Description 00h 01h 02h 03h 04h 05h 06h 07h 08h 09h 0Ah 0Bh 0Ch 0D-FEh FFh Early, pre-2.0 PCI specification devices Mass storage controller Network controller Display controllers Multimedia devices Memory controllers Bridge devices Simple communication controllers Generic system peripherals Input devices Docking stations Processors Serial bus controllers Reserved Device does not fit defined class codes (default) Table 23. Base Class Code 00h: Early, Pre-2.0 Specification Devices Sub-Class Prog I/F Description 00h 01h 00h 00h All currently implemented devices except VGA-compatible devices VGA-compatible devices Table 24. Base Class Code 01h: Mass Storage Controllers Sub-Class Prog I/F Description 00h 01h 02h 03h 04h 80h 00h xxh 00h 00h 00h 00h SCSI controller IDE controller Floppy disk controller IPI controller RAID controller Other mass storage controller Table 25. Base Class Code 02h: Network Controllers Sub-Class Prog I/F Description 00h 00h Ethernet controller AMCC Confidential and Proprietary DS1596 45 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Table 25. Base Class Code 02h: Network Controllers Sub-Class Prog I/F Description 01h 02h 03h 80h 00h 00h 00h 00h Token ring controller FDDI controller ATM controller Other network controller Table 26. Base Class Code 03h: Display Controllers Sub-Class Prog I/F Description 00h 01h 80h 00h 00h 00h RAM memory controller Flash memory controller Other memory controller Table 27. Base Class Code 04h: Multimedia Devices Sub-Class Prog I/F Description 00h 00h 01h 80h 00h 01h 00h 00h VGA-compatible controller 8514 compatible controller XGA controller Other display controller Table 28. Base Class Code 05h: Memory Controllers Sub-Class Prog I/F Description 00h 01h 80h 00h 00h 00h Video device Audio device Other multimedia device AMCC Confidential and Proprietary DS1596 46 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Table 29. Base Class Code 06h: Bridge Devices Sub-Class Prog I/F Description 00h 01h 02h 03h 04h 05h 06h 07h 80h 00h 00h 00h 00h 00h 00h 00h 00h 00h Host/PCI bridge PCI/ISA bridge PCI/EISA bridge PCI/Micro Channel bridge PCI/PCI bridge PCI/PCMCIA bridge NuBus bridge CardBus bridge Other bridge type Table 30. Base Class Code 07h: Simple Communications Controllers Sub-Class Prog I/F Description 00h 00h 01h 02h 00h 01h 02h 00h Generic XT compatible serial controller 16450 compatible serial controller 16550 compatible serial controller Parallel port Bidirectional parallel port ECP 1.X compliant parallel port Other communications device 01h 80h Table 31. Base Class Code 08h: Base System Peripherals Sub-Class Prog I/F Description 00h 00h 01h 02h 00h 01h 02h 00h 01h 02h 00h 01h 00h Generic 8259 PIC ISA PIC EISA PIC Generic 8237 DMA controller ISA DMA controller EISA DMA controller Generic 8254 system timer ISA system timer EISA system timers (2 timers) Generic RTC controller ISA RTC controller Other system peripheral 01h 02h 03h 80h AMCC Confidential and Proprietary DS1596 47 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Table 32. Base Class Code 09h: Input Devices Sub-Class Prog I/F Description 00h 01h 02h 80h 00h 00h 00h 00h Keyboard controller Digitizer (Pen) Mouse controller Other input controller Table 33. Base Class Code 0Ah: Docking Stations Sub-Class Prog I/F Description 00h 01h 02h 10h 40h 00h 00h 00h 00h 00h Intel386™ Intel486™ Pentium™ Alpha™ Co-processor Table 34. Base Class Code 0Bh: Processors Sub-Class Prog I/F Description 00 01h 02h 03h 04h 00h 00h 00h 00h 00h FireWire™ (IEEE 1394) ACCESS.bus SSA Universal Serial Bus Fibre Channel Table 35. Base Class Code 0Ch: Serial Bus Controllers Sub-Class Prog I/F Description 00h 80h 00h 00h Generic docking station Other type of docking station AMCC Confidential and Proprietary DS1596 48 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book CACHE LINE SIZE REGISTER (CALN) Register Name: Address Offset: Power-up value: Boot-load: Attribute: Size: Cache Line Size 0Ch 00h, hardwired not used Read Only 8 bits The cache line configuration register is used by bus masters implementing memory write and invalidate commands. The register defines the cache line size in double word (64-bit) increments. The S5920 is a target device not requiring cache. The register is hardwired to 0. Figure 13. Cache Line Size Register 7 00h Cache Line Size (RO) 0 AMCC Confidential and Proprietary DS1596 49 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book LATENCY TIMER REGISTER (LAT) Register Name: Address Offset: Power-up value: Boot-load: Attribute: Size: Latency Timer 0Dh 00h not used Read Only 8 bits The latency timer defines the minimum amount of time that a bus master can retain ownership of the PCI bus. The S5920 is a target device requiring zero bus ownership time. The register is hardwired to zero. Figure 14. Latency Timer Register 7 00h 0 AMCC Confidential and Proprietary DS1596 50 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book HEADER TYPE REGISTER (HDR) Register Name: Power-up value: Attribute: Size: Header Type Address Offset 0Eh 00h, Hardwired Boot-load: External nvRAM offset 04Eh Read Only 8 bits This register consists of two fields: Bits 6:0 define the format of double words 4d through 15d of the device’s configuration header. Bit 7 defines whether the device is a single function or a multi-function PCI bus agent. The S5920 is defined as a single function PCI device. Figure 15. Header Type Register 7 0 6 00h Configuration header format 0 = single function device 1= multi function device 0 AMCC Confidential and Proprietary DS1596 51 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book BUILT-IN SELF-TEST REGISTER (BIST) Register Name: Power-up value: Boot-load: Attribute: Size: Built-in Self-Test Address Offset 0Fh 00h External nvRAM/EPROM offset 04Fh D7, D5-0 Read Only, D6 as PCI bus write only 8 bits The Built-In Self-Test (BIST) Register permits the implementation of custom, user-specific diagnostics. This register has four fields shown in Figure 10. Bit 7 defines S5920's support of a built-in self test. When bit 7 is set, writing a 1 to bit 6 produce an interrupt signal on the Add-On bus. Bit 6 remains set until cleared by a write operation to this register from the Add-On bus interface. When bit 6 is reset, it is interpreted as completion of the self-test and an error is indicated by a non-zero value for the completion code (bits 3:0). Figure 16. Built-In Self-Test Register 7 X 6 0 5 00 4 3 X Completion Code (RO) Reserved (RO) Start BIST (R/WS) BIST Capable (RO) 0 Bit Description 7 6 BIST Capable. This bit indicates the Add-On device supports a built-in self-test when a 1 is returned. A 0 should be returned if this self-test feature is not required. This field is read only from the PCI interface. Start BIST. Writing a 1 to this bit indicates that the self-test should start. This bit can only be written if bit 7 is one. When bit 6 is set, an interrupt is issued to the Add-On interface. Other than through a reset, Bit 6 can only be cleared by a write to this element from the Add-On bus interface. The PCI bus specification requires that this bit be cleared within 2 seconds after being set, or the device will be failed. This bit is read/write set (R/WS). Reserved. These bits are reserved and are hardwired to 0. Completion Code. This field provides a method for detailing a device-specific error. It is considered valid when the start BIST (bit 6) changes from 1 to 0. An all-zero value for the completion code indicates successful completion. 5:4 3:0 AMCC Confidential and Proprietary DS1596 52 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book BASE ADDRESS REGISTER (BADR) Register Name: Address Offset: Power-up value: Boot-load: Attribute: Size: Determining Base Address Size Base Address 10h, 14h, 18h, 1Ch, 20h FFFFFF81h for offset 10h; 00000000h for all others External nvRAM offset 050h, 54h, 58h, 5Ch, 60h (BADR0-4) high bits Read/Write; low bits Read Only 32 bits The address space defined by a given base address register is determined by writing all 1s to a given base address register from the PCI bus and then reading that register back. The number of 0s returned starting from D4 for memory space and D2 for I/O space toward the high-order bits reveals the amount of address space desired. Tables 17 and 18 list the possible returned values and their corresponding size for both memory and I/O, respectively. Included in the tables are the nvRAM/EPROM boot values which correspond to a given assigned size. A register returning all 0s indicates the region is disabled. Assigning the Base Address Base address registers are used by the system BIOS to determine how much memory or I/O address space a region requires in host space. The actual memory or I/O location(s) of the space is determined by interrogating these registers after BIOS power-up initialization. Bit zero of each field is used to select whether the space required is to be decoded as memory (bit 0 = 0) or I/O (bit 0 = 1). Since this PCI device has internal operating registers, the Base Address Register at offset 10h is assigned to them. The remaining four base address registers can only be used by boot-loading them from the external nvRAM interface. After a base address has been sized, the BIOS can physically locate it in memory (or I/O) space. The base address value must be on a natural binary boundary for the required size. For example, the first base address register returns FFFFFF81h indicating an I/O space (D0=1) of size 80h. This means that the 5920’s internal registers can be selected for I/O addresses between 00000300h through 0000037Fh, in this example. (example 300h, 380h etc.; 338h, 340h would not be allowable). AMCC Confidential and Proprietary DS1596 53 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Figure 17. Base Address Register - Memory 31 x 30 x 29 Base Address 4 3 0 2 0 1 0 0 0 Bit Value Memory Space Indicator (RO) Type (RO) 00 = Locate Anywhere (32) 01 = Below 1 MB 10 = Locate Anywhere (64) 11 = Reserved Prefetchable (RO) Programmable (RO) D31 0 0 1 1 D30 0 1 0 1 Add-On Bus Width Region Disable 8 bits 16 bits 32 bits Bit Description 31:4 Base Address Location. These bits locate the decoded region in memory space. Only bits which return a 1 after being written as 1 are usable for this purpose. Except for Base Address Register 0, these bits are individually enabled by the contents sourced from the external boot memory. Prefetchable. When set as a 1, this bit signifies that this region of memory can be cached. Cacheable regions can only be located within the region altered through PCI bus memory writes. This bit, when set, also implies that all read operations will return the data associated for all bytes regardless of the Byte Enables. Memory space which cannot support this behavior should leave this bit in the zero state. this bit is set by the reset pin and later initialized by the external boot memory option. Base Address Register 0 always has this bit set to 0. This bit is read only from the PCI interface. This bit has no implementation in the S5920 other than providing it during a configuration read cycle. Memory Type. These bits define whether the memory space is 32 or 64 bits wide and if the space location is restricted to be within the first megabyte of memory space. The encoding is as follows: Bits Description 2 1 3 2:1 0 0 1 1 2 0 1 0 1 Region is 32-bits wide and can be located anywhere in 32-bit memory space. Region is 32 bits wide and must be mapped below the first Mbytes of memory space. Region is 64 bits wide and can be mapped anywhere within 64-bit memory space. (Not supported by this device.) Reserved. Note: The 64-bit memory space is not supported by this device. Bit 2 is hardwired to 0. Options are restricted to desired to memory space anywhere within 32-bit memory space or located in the first megabyte. For Base Addresses 1 through 4, this bit is cleared by the reset pin and later initialized by the external boot memory option. Space Indicator = 0. When set to 0, this bit defines a base address region as memory space and the remaining bits in the base address register are defined as shown in Figure 17. 0 AMCC Confidential and Proprietary DS1596 54 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Figure 18. Base Address Register - I/O 31 Base Address 2 1 0 0 1 Bit Value I/O Space Indicator (RO) Reserved (RO) Programmable (R/W) Bit Description 31:2 Base Address Location. These bits are used to position the decoded region in I/O space. Only bits which return a 1 after being written as 1 are usable for this purpose. Except for Base Address 0, these bits are individually enabled by the contents sourced from the external nvRAM. Reserved. This bit should be 0. (Note: disabled Base Address Registers will return all 0s for the entire register location, bits 31 through 0). Space Indicator = 1. When 1, this bit identifies a base address region as an I/O space and the remaining bits in the base address register have the definition as shown in Figure 18. 1 0 AMCC Confidential and Proprietary DS1596 55 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Table 36. Base Address Register Response (Memory Assigned) to All-Ones Write Operation Response Size in Bytes nvRAM boot value1 00000000h FFFFFFF0h FFFFFFE0h FFFFFFC0h FFFFFF80h FFFFFF00h FFFFFE00h FFFFFC00h FFFFF800h FFFFF000h FFFFE000h FFFFC000h FFFF8000h FFFF0000h FFFE0000h FFFC0000h FFF80000h FFF00000h FFE00000h FFC00000h FF800000h FF000000h FE000000h FC000000h F8000000h F0000000h E0000000h none - disabled 16 bytes (4 DWORDs) 32 bytes (8 DWORDs) 64 bytes (16 DWORDs) 128 bytes (32 DWORDs) 256 bytes (64 DWORDs) 512 bytes (128 DWORDs) 1K bytes (256 DWORDs) 2K bytes (512 DWORDs) 4K bytes (1K DWORDs) 8K bytes (2K DWORDs) 16K bytes (4K DWORDs) 32K bytes (8K DWORDs) 64K bytes (16K DWORDs) 128K bytes (32K DWORDs) 256K bytes (64K DWORDs) 512K bytes (128K DWORDs) 1M bytes (256K DWORDs) 2M bytes (512K DWORDs) 4M bytes (1M DWORDs) 8M bytes (2M DWORDs) 16M bytes (4M DWORDs) 32M bytes (8M DWORDs) 64M bytes (16M DWORDs) 128M bytes (32M DWORDs) 256M bytes (64M DWORDs) 512M bytes (128M DWORDs) 00000000h or BIOS missing2 FFFFFFF0h FFFFFFE0h FFFFFFC0h FFFFFF80h FFFFFF00h FFFFFE00h FFFFFC00h FFFFF800h FFFFF000h FFFFE000h FFFFC000h FFFF8000h FFFF0000h FFFE0000h FFFC0000h FFF80000h FFF00000h FFE00000h FFC00000h FF800000h FF000000h FE000000h FC000000h F8000000h F0000000h E0000000h 1. The two most significant bits define bus width for BADR1:4 in Pass-Thru operation. (See S5920 Base Address Register Definition.) 2. Bits D3, D2 and D1 may be set to indicate other attributes for the memory space. AMCC Confidential and Proprietary DS1596 56 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Table 37. Read Response (I/O Assigned) to an All-Ones Write Operation to a Base Address Register Response Size in Bytes nvRAM boot value 00000000h FFFFFFFDh FFFFFFF9h FFFFFFF1h FFFFFFE1h FFFFFFC1h FFFFFF81h FFFFFF01h none - disabled 4 bytes (1 DWORDs) 8 bytes (2 DWORDs) 16 bytes (4 DWORDs) 32 bytes (8 DWORDs) 64 bytes (16 DWORDs) 128 bytes (32 DWORDs) 256 bytes (64 DWORDs) 00000000h or BIOS missing FFFFFFFDh FFFFFFF9h FFFFFFF1h FFFFFFE1h FFFFFFC1h FFFFFF81h 1 FFFFFF01h 1. Base Address Register 0, at offset 10h, powers up as FFFFFF81h. This default assignment allows usage without an external boot memory. Should an nvRAM be used, the base address can be boot loaded to become a memory space (FFFFFF80h or FFFFFF82h). AMCC Confidential and Proprietary DS1596 57 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book SUBSYSTEM VENDOR IDENTIFICATION REGISTER (SVID) Register Name: Address Offset: Power-up value: Boot-load: Attribute: Size: Subsystem Vendor ID 2Ch-2Dh 0000h External nvRAM offset 6Ch-6Dh Read Only (RO) 16 bits This register is used to uniquely identify the user board or subsystem. It provides a mechanism for add-in card vendors to distinguish devices with the same Vendor ID and Device ID. Implementation of this register is mandatory for 2.2 compliance and an all-zero value indicates that the device does not support subsystem identification. Subsystem Vendor IDs may be obtained directly from the PCI SIG by the user and it is loaded by the S5920 from the external nvRAM at power up. Figure 19. Subsystem Vendor Identification Register 15 0000h 0 Subsystem Vendor Identification Register (RO) Bit Description 15:0 Subsystem Vendor Identification Number. AMCC Confidential and Proprietary DS1596 58 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book SUBSYSTEM ID REGISTER (SID) Register Name: Address Offset: Power-up value: Boot-load: Attribute: Size: Subsystem Identification 2Eh-2Fh 0000h External nvRAM offset 6Eh-6Fh Read Only (RO) 16 bit This register is used to further identify the add-in board or subsystem. It provides a mechanism for add-in card vendors to distinguish devices with the same Vendor ID and Device ID. Implementation of this register is mandatory for 2.2 compliance and an all-zero value indicates that the device does not support subsystem identification. Subsystem ID is vendor-specific. It is loaded by the S5920 from the external nvRAM at power up. Figure 20. Subsystem Identification Register 15 0000h 0 Subsystem Identification Register (RO) Bit Description 15:0 Subsystem Identification Number. AMCC Confidential and Proprietary DS1596 59 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book EXPANSION ROM BASE ADDRESS REGISTER (XROM) Register Name: Address Offset: Power-up value: Boot-load: Attribute: Size: Expansion ROM Base Address 30h 00000000h External nvRAM offset 70h bits 31:11, bit 0 Read/Write; bits 10:1 Read Only 32 bits The Expansion ROM Base Address Register provides a mechanism for assigning a space within physical memory for a BIOS expansion ROM. Access from the PCI bus to the memory space defined by this register will cause one or more accesses to the S5920 external nvRAM interface. Since PCI bus accesses to the ROM may be 32 bits wide, repeated read operations to the ROM are generated, and the wider data is assembled internal to the S5920 controller. The data is then transferred to the PCI bus by the S5920. Only memory read cycles should be performed to this location. Figure 21. Expansion ROM Base Address Register 31 11 10 0 1 0 0 Bit Value Address Decode Enable (RW) 0 = Disabled 1 = Enabled Reserved (RO) Programmable (R/W) Bit Description 31:11 Expansion ROM Base Address Location. These bits are used to position the decoded region in memory space. Only bits which return a 1 after being written as 1 are usable for this purpose. These bits are individually enabled by the contents sourced from the external boot memory (nvRAM). The desired size for the ROM memory is determined by writing all ones to this register and then reading back the contents. The number of bits returned as zeros, in order from least significant to most significant bit, indicates the size of the expansion ROM. This controller limits the expansion ROM area to 2K bytes (due to the serial nvRAMÕs limit of 11 bits of address). The allowable returned values after all ones are written to this register are shown in Table 18. Reserved. All zeros. Address Decode Enable. The Expansion ROM address decoder is enabled or disabled with this bit. When this bit is set, the decoder is enabled. When this bit is cleared, the decoder is disabled. It is requiredt the PCI command register (PCICMD) also have the memory decode bit enabled for this bit to have any effect. In addition, the corresponding bit must be set in the external nvRAM (see page 2-74, Table 1). If not set, the PCI host cannot enable/disable this Address Decode bit. 10:1 0 AMCC Confidential and Proprietary DS1596 60 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Table 38. Read Response to Expansion ROM Base Address Register (after all ones written) Response Size in Bytes nvRAM boot value 00000000h FFFFF801h none - disabled 2K bytes (512 DWORDs) 00000000h or BIOS missing 1 FFFFF801h 1. The Expansion ROM Base Address Register nvRAM boot value is internally hardwired to FFFFF80Xh, where X = 000xb (i.e., only the leastsignificant bit, or Address Decode Enable bit, is programmable). This defines both the minimum and maximum expansion ROM size supported by the S5920 (2K bytes). The Address Decode Enable bit in the nvRAM (the LSB) must be set to enable this region. If not set, a PCI Configuration read of this region will always respond with 00000000h. AMCC Confidential and Proprietary DS1596 61 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book INTERRUPT LINE REGISTER (INTLN) Register Name: Address Offset: Power-up value: Boot-load: Attribute: Size: Interrupt Line 3Ch FFh External nvRAM offset 7Ch Read/Write 8 bits This register indicates the interrupt routing for the S5920 controller. The ultimate value for this register is system-architecture specific. For x86 based PCs, the values in this register correspond with the established interrupt numbers associated with the dual 8259 controllers used in those machines. In x86-based PC systems, the values of 0 to 15 correspond with the IRQ numbers 0 through 15, and the values from 16 to 254 are reserved. The value of 255 (the controller’s default power-up value) signifies either “unknown” or “no connection” for the system interrupt. This register is bootloaded from the external boot memory or may be written by the PCI interface. Figure 22. Interrupt Line Register 7 FFh 0 AMCC Confidential and Proprietary DS1596 62 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book INTERRUPT PIN REGISTER (INTPIN) Register Name: Address Offset: Power-up value: Boot-load: Attribute: Size: Interrupt Pin 3Dh 01h External nvRAM offset 7Dh Read Only 8 bits This register identifies which PCI interrupt, if any, is connected to the controller’s PCI interrupt pins. The allowable values are 0 (no interrupts), 1 (INTA#), 2 (INTB#), 3 (INTC#), and 4 (INTD#). The default powerup value assumes INTA#. Figure 23. Interrupt Pin Register 7 00h 3 2 X 1 X 0 X Bit Value Pin Number 000 None 001 INTA# 010 INTB# 011 INTC# 100 INTD# 101 Reserved 11X Reserved Reserved (All Zeros - RO) AMCC Confidential and Proprietary DS1596 63 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book MINIMUM GRANT REGISTER (MINGNT) Register Name: Address Offset: Power-up value: Boot-load: Attribute: Size: Minimum Grant 3Eh 00h, hardwired not used Read Only 8 bits This register may be optionally used by bus masters to specify how long a burst period the device needs. A value of zero indicates that the bus master has no stringent requirement. The units defined by the least significant bit are in 250 ns increments. This register is treated as “information only” since the S5920 is a PCI target device only. Figure 24. Minimum Grant Register 7 00h 0 Bit Value Value x 250ns (RO) 00 - No Requirement 01 - FFh AMCC Confidential and Proprietary DS1596 64 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book MAXIMUM LATENCY REGISTER (MAXLAT) Register Name: Address Offset: Power-up value: Boot-load: Attribute: Size: Maximum Latency 3Fh 00h, hardwired not used Read Only 8 bits This register may be optionally used by bus masters to specify how often this device needs PCI bus access. A value of zero indicates that the bus master has no stringent requirement. The units defined by the least significant bit are in 250 ns increments. Since the S5920 is a PCI target device only, this register is treated as “information only” and has no further implementation within this device. Figure 25. Maximum Latency Register 7 00h 0 Bit Value AMCC Confidential and Proprietary DS1596 65 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book OPERATION REGISTERS All S5920 control and communications are performed through two register groups: PCI Operation Registers Add-On Operation Registers. Some registers in both groups are accessible from both buses. This chapter describes the PCI Operation Register set first and then the Add-On Operation Register set for easier understanding. An access to a register common to both buses at the same time is not allowed. Unpredictable behavior may occur. Table 39. Operation Registers - PCI Bus Address Offset Abbreviation Register Name PCI BUS OPERATION REGISTERS The PCI bus operation registers are mapped as 6 DWORD registers located at the address space (I/O or memory) specified by Base Address Register 0. These locations are the primary method of communication between the PCI and Add-On buses. It is NOT recommended to read or write from an undefined address. The read results and write effects cannot be guaranteed. Table 1 lists the PCI Bus Operation Registers. 0Ch 1Ch 34h 38h 3Ch 60h OMB IMB MBEF INTCSR RCR PTCR Outgoing Mailbox Register Incoming Mailbox Register Mailbox Empty/Full Status Register Interrupt Control/Status Register Reset Control Register Pass-Thru Configuration Register Note: Absolute register address locations are acquired by adding BADR0 to the “address offset” listed above. AMCC Confidential and Proprietary DS1596 66 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book OUTGOING MAILBOX REGISTER (OMB) Register Names: PCI Address Offset: Power-up value: PCI Attribute: Size: Outgoing Mailbox 0Ch Undefined Read/Write 32 bits This DWORD register provides a method for sending command or parameter data to the Add-On bus. PCI bus transactions to this register may be of any width (byte, word, or DWORD). Writing to this register can be a source for Add-On bus interrupts by enabling interrupt generation through the use of the Add-on’s Interrupt Control/Status Register. This is also called the Add-On Incoming Mailbox Register (AIMB). Reading from this register will not affect interrupts or the MBEF status register. Figure 26. Outgoing Mailbox 31 Byte 3 24 23 Byte 2 16 15 Byte 1 8 7 Byte 0 0 AMCC Confidential and Proprietary DS1596 67 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book PCI INCOMING MAILBOX REGISTER (IMB) Register Names: PCI Address Offset: Power-up value: PCI Attribute: Size: Incoming Mailbox 1Ch Undefined Read Only 32 bits This DWORD register provides a method for receiving user-defined status or parameter data from the AddOn bus. PCI bus transactions to this register may be of any width (byte, word, or DWORD). Only read operations are supported from this register. Reading from this register can be a source for Add-On bus interrupt by enabling interrupt generation through the use of the Add-On Interrupt Control/Status Register. Byte 3 of this mailbox can also be controlled via external hardware from the Add-On bus. This register is also referred to as the Add-On Outgoing Mailbox Register AOMB). Figure 27. Incoming Mailbox 31 Byte 3 24 23 Byte 2 16 15 Byte 1 8 7 Byte 0 0 AMCC Confidential and Proprietary DS1596 68 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book PCI MAILBOX EMPTY/FULL STATUS REGISTER (MBEF) Register Name: PCI Address Offset: Power-up value: PCI Attribute: Size: Mailbox Empty/Full Status 34h 00000000h Read Only 32 bits This register provides empty/full visibility for each byte within the mailboxes. The empty/full status for the PCI Outgoing mailbox is displayed on bits 15 to 12 and the empty/full status for the PCI Incoming mailbox is presented on bits 31 to 28. A value of 1 signifies that a given mailbox has been written by one bus interface but has not yet been read by the corresponding destination interface. The PCI bus incoming mailbox transfers data from the Add-On bus to the PCI bus, and the PCI outgoing mailbox transfers data from the PCI bus to the Add-On bus. This register is also referred to as the Add-On Mailbox Empty/Full Status Register (AMBEF). Figure 28. Mailbox Empty/Full Status Register (MBEF) 31 28 27 0000 Reserved 16 15 12 11 0000 0 Reserved PCI Outgoing Mailbox Status (RO) PCI Incoming Mailbox Status (RO) Table 40. Mailbox Empty/Full Status Register Bit Description 31:28 PCI Incoming Mailbox Status. This field indicates which byte of the incoming mailbox register has been written by the Add-On interface but has not been read by the PCI bus. Each bit location corresponds to a specific byte within the incoming mailbox. A value of one for each bit signifies that the specified mailbox byte is full, and a value of 0 signifies empty. The mapping of these status bits to bytes within the mailbox is as follows: Bit 31 = Incoming mailbox byte 3 Bit 30 = Incoming mailbox byte 2 Bit 29 = Incoming mailbox byte 1 Bit 28 = Incoming mailbox byte 0 PCI Outgoing Mailbox Status. This field indicates which byte of the outgoing mailbox register has been written by the PCI bus interface but has not yet been read by the Add-On bus. Each bit location corresponds to a specific byte within the outgoing mailbox. A value of one for each bit signifies that the specified mailbox byte is full, and a value of 0 signifies empty. The mapping of these status bits to bytes is as follows: Bit 15 = Outgoing mailbox byte 3 Bit 14 = Outgoing mailbox byte 2 Bit 13 = Outgoing mailbox byte 1 Bit 12 = Outgoing mailbox byte 0 15:12 AMCC Confidential and Proprietary DS1596 69 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book PCI INTERRUPT CONTROL/STATUS REGISTER (INTCSR) Register Name: PCI Address Offset: Power-up value: PCI Attribute: Size: Interrupt Control and Status 38h 00000C0Ch Read/Write, Read/Write Clear 32 bits This register configures the conditions which will produce an interrupt on the PCI bus interface, a method for viewing the cause of the interrupt, and a method for acknowledging (removing) the interrupt’s assertion. Interrupt sources: • • • The Outgoing mailbox becomes empty. The Incoming mailbox becomes full. Add-On interrupt pin enable and flag. Figure 29. Interrupt Control Status Register Actual Interrupt 3 1 2 4 2 3 2 2 2 1 2 0 1 9 1 8 1 7 1 6 1 5 1 4 1 3 1 2 1 1 1 0 Interrupt Selection 9 XX 8 75 000 432 X11 1 XX 0 00000000 XX0000XX00XX11 Interrupt Asserted (RO) Add-On Interrupt Pin (ADDINT#) Status (RO) Incoming Mailbox Interrupt (R/WC) Outgoing Mailbox Interrupt (R/WC) Add-On Interrupt Pin (ADDINT#) Enable (R/W) Interrupt Source (R/W) Enable & Selection D4 - D0 PCI Outgoing Mailbox (Becomes Empty) D4 = Enable Interrupt D1 - D0 = Byte Number 00 = Byte 0 01 = Byte 1 D8 - D12 PCI Incoming Mailbox 10 = Byte 2 D12 = Enable Interrupt 11 = Byte 3 D9 - D8 = Byte Number 00 = Byte 0 01 = Byte 1 10 = Byte 2 11 = Byte 3 AMCC Confidential and Proprietary DS1596 70 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Table 41. Interrupt Control Status Register Bit Description 31:24 23 Reserved. Always zero. Interrupt Asserted. This read only status bit indicates that one or more of the three possible interrupt conditions are present. This bit is the OR of the mailbox interrupt conditions described by Bits 17 and 16, as well as the OR of the Add-On interrupt described in Bit 22 (if the Add-On Interrupt is Enabled with Bit 13). No PCI interrupt is generated, nor is this bit ever set, for an Add-On Interrupt without the Add-On Interrupt Enable set. Add-On Interrupt. This bit is set when the ADDINT# input pin is driven low by an Add-On bus device. A high bit indicates an Add-On device is requesting service. In addition, if the ADDINT# Enable bit is set, the S5920 will assert a PCI interrupt (INTA# driven low). The source driving ADDINT# must deassert this input before the PCI interrupt (INTA#) is driven to a false state. Host software must clear the Add-On interrupt source before exiting its interrupt handler routine. Reserved. Always zero. "PCI Incoming Mailbox Interrupt. This bit can be set when the mailbox is written by the Add-On interface. This bit operates as read or write one clear. A write to this bit with the data of ""one"" will cause this bit to be reset; a write to this bit with the data of ""O "" will not change the state of this bit." "PCI Outgoing Mailbox Interrupt. This bit can be set when the mailbox is read by the Add-On interface. This bit operates as read or write one clear. A write to this bit with the data of ""one"" will cause this bit to be reset; a write to this bit with the data of ""O "" will not change the state of this bit." Reserved. Always zero. ADDINT# Enable. If this bit is high, the S5920 will allow the Add-On interrupt request to drive the INTA# pin. It has no effect on the assertion of the Add-On Interrupt Bit 22. Enable Incoming Mailbox interrupt. This bit allows a write from the incoming mailbox register byte identified by bits 9 and 8 to produce a PCI interface interrupt. This bit is read/write. Hardwired to 1. Reserved. Incoming Mailbox Byte Interrupt Select. This field selects which byte of the mailbox is to actually cause the interrupt. [00]b selects byte 0, [01]b selects byte 1, [10]b selects byte 2, and [11]b selects byte 3. This field is read/ write. Reserved. Always zero. Enable Outgoing Mailbox Interrupt. This bit allows a read by the Add-On of the outgoing mailbox register byte identified by bits 1 and 0 to produce a PCI interface interrupt. This bit is read/write. Hardwired to 1. Reserved. Outgoing Mailbox Byte Interrupt Select. This field selects which byte of the mailbox is to actually cause the interrupt. [00]b selects byte 0, [01]b selects byte 1, [10]b selects byte 2, and [11]b selects byte 3. This field is read/write. 22 21:18 17 16 15:14 13 12 11:10 9:8 7:5 4 3:2 1:0 AMCC Confidential and Proprietary DS1596 71 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book PCI RESET CONTROL REGISTER (RCR) Register Name: PCI Address Offset: Power-up value: Attribute: Size: Reset Control Register 3Ch 00000000h Read/Write, Read Only, Write Only 32 bits This register provides a method to perform software resets. It will also control nvRAM accesses. The following controls are available: • • • • Assert reset to Add-On Reset mailbox empty full status flags Write/Read external non-volatile memory Reset Pass-Thru Read FIFO Figure 30. FIFO Control/Status Register 2 8 2 7 2 6 2 5 2 4 31 29 23 16 15 00000000000 0 0 nvRAM Operation Address/Data (R/W) Add-On Reset (R/W) Read FIFO Reset (WO) Mailbox Flags Reset (WO) nvRAM Access Failed (RO) nvRAM Access Control AMCC Confidential and Proprietary DS1596 72 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Table 42. Reset Control Register Bit Description 31:29 nvRAM Access Control. This field provides a method for access to the optional external non-volatile memory. Write operations are achieved by a sequence of byte operations involving these bits and the 8-bit field of bits 23 through 16. The sequence requires that the low-order address, high-order address, and then a data byte are loaded in order. Bit 31 of this field acts as a combined enable and ready for the access to the external memory. D31 must be written to a 1 before an access can begin, and subsequent accesses must wait for bit D31 to become 0 (ready). D31 0 1 1 1 1 0 1 D30 X 0 0 1 1 X X D29 X 0 1 0 1 X X W/R W W W W W R R Inactive Load low address byte Load high sddress byte Begin write Begin read Ready Busy Cautionary note: The nonvolatile memory interface is also available for access by the Add-On interface. While simultaneous accesses to the nv memory by both the Add-On and PCI are supported (via arbitration logic), software must be designed to prevent the possibility of data corruption within the memory and to provide for accurate data retrieval. 28 27 nvRAM Access Failed. Indicate the last nvRAM access failed. This flag is cleared automatically upon the start of the next read/write operation. Mailbox Flag Reset. Writing a one to this bit causes all mailbox status flags to become reset (EMPTY). It is not necessary to write this bit to 0 afterwards because it is used internally to produce a reset pulse. Since reading this bit will always return a 0, this bit is write only. Reserved. Always zero. Read FIFO Reset. Writing a one to this bit causes the read FIFO to reset (empty). It is not necessary to write a 0 to this bit. This bit is write only. This feature is intended for test only. However, it can be used during operation if several PCI idle cycles are inserted following the assertion of this command. Add-On Pin Reset. Writing a one to this bit causes the reset output pin to become active (SYSRST#). Clearing this bit is necessary in order to remove the assertion of reset. This bit is read/write. Non-volatile Memory Address/Data Port. This 8-bit field is used in conjunction with bits 31, 30 and 29 of this register to access the external non-volatile memory. The contents written are either low address, high address, or data as defined by bits 30 and 29. This register will contain the external non-volatile memory data when the proper read sequence for bits 31 through 29 is performed. Reserved. Always zero. 26 25 24 23:16 15:0 AMCC Confidential and Proprietary DS1596 73 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book PCI PASS-THRU CONFIGURATION REGISTER (PTCR) Register Name: PCI Address Offset: Power-up value: PCI Attribute: Size: This register controls the configuration for Pass-Thru Regions 1-4: Byte 0 Controls Pass-Thru Region 1 Byte 1 Controls Pass-Thru Region 2 Byte 2 Controls Pass-Thru Region 3 Byte 3 Controls Pass-Thru Region 4 IMPORTANT NOTE: This register (PTCR) is physically the same as the Add-On Pass-Thru Configuration Register (APTCR). It is intended that either the PCI system or local Add-On interface will write to this register, but not both. However, in the event that both the PCI and Add-On must write to this register, whichever side wrote last will update its value. Pass-Thru Configuration Register 60h 80808080h Read/Write 32 bits Also, Pass-Thru operation cannot be guaranteed if this register is updated while a Pass-Thru operation is already in progress. Figure 31. Pass-Thru Configuration Register 31 Region 4 24 23 Region 3 16 15 Region 2 87 Region 1 0 D7 Pass-Thru Mode 0 - PTADR# Not Driven 1 - PTADR# Driven D6 Endian Conversion 0 - No Conversion 1 - Big Endian Conversion D5 Write FIFO 0 - Enabled 1 - Disabled Active Mode Wait States D2 D1 D0 000 - 0 Wait States 001 - 1 Wait States 010 - 2 Wait States 011 - 3 Wait States 100 - 4 Wait States 101 - 5 Wait States 110 - 6 Wait States 111 - 7 Wait States Prefetch D4 D3 00 - Disabled 01 - Small 10 - Medium 11 - Large AMCC Confidential and Proprietary DS1596 74 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Table 5 describes one of the four configuration registers. All four region configuration registers are exactly the same. Table 43. Bit Description 7 PTADR# mode. This bit is only valid in Active mode. If this bit is 0 , PTADR# is not driven at the beginning of an active cycle. If this bit is set to 1 (default state), the S5920 will assert PTADR# for one clock cycle after PTATN# is asserted. The Pass-Thru address is also driven while PTADR# is low. This bit is a donÕt care if the device is operating in Passive mode. Endian conversion. If this bit is set to one, the S5920 will convert the Add-On bus from the default little endian format to a big endian format. Reference Chapter 9 for more details. Write FIFO disabled. If this bit is set to 1, the S5920 will not accept the next piece of data (on a PCI write) until the Add-On has accepted the previous piece of data. If this bit is set to 0, the S5920 will accept data from the PCI until the Pass-Thru write FIFO is full. Prefetch. These bits control the number of DWORDs the S5920 will prefetch after the current PCI Pass-Thru read completes. The actual amount of data prefetched depends upon any number of different scenarios. The prefetch values of ÒsmallÓ, ÒmediumÓ and ÒlargeÓ are available to tune the system to achieve best overall performance (i.e., optimize PCI bus transfers or optimize Add-On bus transfers). The Pass-Thru read FIFO can be enabled to prefetch in either Active mode or Passive mode. Wait states. In Active mode, the user can program the number of wait states required by the Add-On bus to complete a transaction. Up to 7 wait states can be programmed (per region). The S5920 will count the number of clocks programmed into this register before finishing the current data transaction if PTWAIT# is high. If PTWAIT# is driven low, additional wait states may be inserted. Bits 2, 1 and 0 are donÕt care if operating in Passive mode. 6 5 4:3 2:0 AMCC Confidential and Proprietary DS1596 75 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book ADD-ON BUS OPERATION REGISTERS The Add-On bus interface provides access to 8 DWORDs of data, control and status information. All of these locations are accessed by asserting the Add-On bus chip select pin (SELECT#) and the byte-enable pins (BE[3:0]), in conjunction with either the read or write control enables (signal pin RD# or WR#). All registers are accessed with signals synchronous to the Add-On clock. Table 44. Operation Registers - Add-On Interface Address Abbreviation Register Name This register group represents the primary method for communication between the Add-On and PCI buses as viewed by the Add-On. The flexibility of this arrangement allows a number of user-defined software protocols to be built. One should NOT read/write from any undefined address, or the read results and write effects cannot be guaranteed. Table 6 lists the Add-On Bus Operation Registers. 0Ch 1Ch 28h 2Ch 34h 38h 3Ch 60h AIMB AOMB APTA APTD AMBEF AINT ARCR APTCR Add-On Incoming Mailbox Register Add-On Outgoing Mailbox Register Add-On Pass-Thru Address Register Add-On Pass-Thru Data Register Add-On Mailbox Empty/Full Status Register Add-On Interrupt Control/Status Register Add-On Reset Control Register Add-On Pass-Thru Configuration Register AMCC Confidential and Proprietary DS1596 76 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book ADD-ON INCOMING MAILBOX REGISTER (AIMB) Register Names: Incoming Mailbox Add-On Address: Power-up value: Add-On Attribute: Size: 0Ch XXXXXXXXh Read Only 32 bits This DWORD register provides a method for receiv-ing user-defined status or parameter data from the PCI system. Add-On bus read operations to this register may be of any width (byte, word, or DWORD). Only read operations are supported. Reading from this register can optionally cause a PCI bus interrupt (if desired) by enabling interrupt generation through the use of the PCI’s Interrupt Control/Status Register. This register is also referred to as the PCI Outgoing Mailbox Register. ADD-ON OUTGOING MAILBOX REGISTER (AOMB) Register Names: Add-On Address: Power-up value: Add-On Attribute: Size: Outgoing Mailbox 1Ch XXXXXXXXh Read/Write 32 bits This DWORD register provides a method for sending command or parameter data to the PCI interface. Add-On bus operations to this register may be of any width (byte, word, or DWORD). Writing to this register can be a source for PCI bus interrupts (if desired) by enabling interrupt generation through the use of the PCI’s Interrupt Control/Status Register. This is also called the PCI Incoming Mailbox Register (IMB). Byte 3 of this mailbox can also be controlled via the external mailbox port. Reading from this register will not affect interrupts or the AMBEF Status Register. (OMB). ADD-ON PASS-THRU ADDRESS REGISTER (APTA) Register Name: Add-On Address: Power-up value: Add-On Attribute: Size: Add-On Pass-Thru Address 28h XXXXXXXXh Read Only 32 bits This register stores the address of any active Pass-Thru PCI bus cycle that has been accepted by the S5920. When one of the base address decode registers 1-4 encounters a PCI bus cycle which selects the region defined by it, this register stores that current cycle’s active address. This address is incremented after every 32-bit Pass-Thru data transfer. ADD-ON PASS-THRU DATA REGISTER (APTD) Register Name: Add-On Address: Power-up value: Add-On Attribute: Size: Add-On Pass-Thru Data 2Ch XXXXXXXXh Read/Write 32 bits This register, along with APTA register, is used to perform Pass-Thru transfers. When one of the base address decode registers 1-4 encounters a PCI bus cycle which selects the region defined by it, the APTA register will contain that current cycle’s active address and the APTD will contain the data (PCI bus writes) or must be written with data (PCI bus reads). Wait states are generated on the PCI bus until this register is read (PCI bus writes) or this register is written (PCI bus reads) when in Passive mode. AMCC Confidential and Proprietary DS1596 77 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book ADD-ON MAILBOX EMPTY/FULL STATUS REGISTER (AMBEF) Register Name: Add-On Address: Power-up value: Add-On Attribute: Size: Mailbox Empty/Full Status 34h 00000000h Read Only 32 bits This register provides empty/full visibility for each byte within the mailboxes. The empty/full status for the Add-On Incoming mailbox is displayed on bits 15 to 12 and the empty/full status for the Add-On Out-going mailbox is presented on bits 31 to 28. A value of 1 signifies that a given mailbox has been written by one bus interface but has not yet been read by the corresponding destination interface. The Add-On bus incoming mailbox is used to transfer data from the PCI bus to the Add-On bus, and the Add-On outgoing mailbox is used to transfer data from the Add-On bus to the PCI bus. This register is also referred to as the PCI Mailbox Empty/Full Status Register (MBEF). Figure 32. Mailbox Empty/Full Status Register 31 28 27 16 0000 15 12 11 0000 0 Reserved Add-On Outgoing Mailbox Status (RO) Reserved Add-On Incoming Mailbox Status (RO) AMCC Confidential and Proprietary DS1596 78 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Table 45. Mailbox Empty/Full Status Register Bit Description 31:28 Add-On Outgoing Mailbox Status. This field indicates which byte of the outgoing mailbox register has been written by the Add-On interface but has not yet been read by the PCI bus. Each bit location corresponds to a specific byte within the outgoing mailbox. A value of 1 for each bit signifies that the specified mailbox byte is full, and a value of 0 signifies empty. The mapping of these status bits to bytes within the mailbox is as follows: Bit 31 = Outgoing mailbox byte 3 Bit 30 = Outgoing mailbox byte 2 Bit 29 = Outgoing mailbox byte 1 Bit 28 = Outgoing mailbox byte 0 Add-On Incoming Mailbox Status. This field indicates which byte of the incoming mailbox register has been written by the PCI bus interface but has not yet been read by the Add-On bus. Each bit location corresponds to a specific byte within the incoming mailbox. A value of 1 for each bit signifies that the specified mailbox byte is full, and a value of 0 signifies empty. The mapping of these status bits to bytes is as follows: Bit 15 = Incoming mailbox byte 3 Bit 14 = Incoming mailbox byte 2 Bit 13 = Incoming mailbox byte 1 Bit 12 = Incoming mailbox byte 0 15:12 AMCC Confidential and Proprietary DS1596 79 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book ADD-ON INTERRUPT CONTROL/STATUS REGISTER (AINT) Register Name: Add-On Address: Power-up value: Attribute: Size: Add-On Interrupt Control and Status 38h 00000C0Ch Read/Write, Read/Write Clear 32 bits This register provides the method for choosing which conditions are to produce an interrupt on the Add-On bus interface, a method for viewing the cause for the interrupt, and a method for acknowledging (removing) the interrupt’s assertion. Interrupt sources: • • • Incoming mailbox becomes full Outgoing mailbox becomes empty Built-in self test issued Figure 33. Add-On Interrupt Control Status Register Interrupt Status Interrupt Selection 2 0 1 9 1 8 1 7 1 6 1 5 1 4 1 3 1 2 1 1 1 0 31 0000 24 2 3 2 2 2 1 8 XX 000 432 X11 XX 0 X00X00XX000 11 Interrupt Asserted (RO) BIST (R/WC) Outgoing Mailbox Interrupt (R/WC) Incoming Mailbox Interrupt (R/WC) Add-On IMB (Becomes full) R/W D4 = Enable Interrupt D1 - D0 = Byte Number 00 = Byte 0 01 = Byte 1 Add-On OMB (Becomes Empty) 10 = Byte 2 R/W 11 = Byte 3 D12 = Enable Interrupt D9 - D8 = Byte Number 00 = Byte 0 01 = Byte 1 10 = Byte 2 11 = Byte 3 AMCC Confidential and Proprietary DS1596 80 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Table 46. Interrupt Control Status Register Bit Description 31:24 23 22:21 20 Reserved. Always zero. Interrupt Asserted. This read-only status bit indicates that one or more interrupt conditions are present. This bit is the OR of the interrupt sources described by bits 20, 17 and 16 of this register. Reserved. Always zero. BIST. Built-In Self-Test Interrupt. This interrupt occurs when a self test is initiated by the PCI interface by writing to the PCI configuration register BIST. This bit will stay set until cleared by writing a 1 to this location. Self test completion codes may be passed to the PCI BIST register by writing to the ARCR register. Reserved. Always zero. Outgoing Mailbox Interrupt. This bit can be set when the mailbox is read by the PCI interface. This bit operates as read or write 1 clear. A write with the data as 1 will cause this bit to be reset; a write with the data as 0 will not change the state of this bit. Incoming Mailbox Interrupt. This bit can be set when the mailbox is written by the PCI interface. This bit operates as read or write 1 clear. A write with the data as 1 will cause this bit to be reset; a write with the data as 0 will not change the state of this bit. Reserved. Always zero. Enable Outgoing Mailbox Interrupt. This bit allows a PCI read of the outgoing mailbox register to produce an Add-On interrupt. This bit is read/write. Hardwired to 11. Outgoing Mailbox Byte Interrupt Select. This field selects which byte of the mailbox is to cause the interrupt. [00]b selects byte 0, [01]b selects byte 1, [10]b selects byte 2, and [11]b selects byte 3. This field is read/write. Reserved. Always zero. Enable Incoming Mailbox Interrupt. This bit allows a write from the PCI bus to the incoming mailbox register to produce an Add-On interrupt. This bit is read/write. Hardwired to 1. Incoming Mailbox Byte Interrupt Select. This field selects which byte of the mailbox is to cause the interrupt. 00b selects byte 0, 01b selects byte 2, and 11b selects byte 3. This field is read/write. 19:18 17 16 15:13 12 11:10 9:8 7:5 4 3:2 1:0 AMCC Confidential and Proprietary DS1596 81 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book ADD-ON RESET CONTROL REGISTER (ARCR) Register Name: Add-On Address: Power-up value: Attribute: Size: This register provides a method to perform software resets and nvRAM accesses. The following Add-On controls are provided: • • • Reset mailbox empty full status flags Reset Pass-Thru read FIFO Read/Write external non-volatile memory Add-On Reset Control and Status 3Ch 00h Read/Write, Read Only, Write Only 32 bits Figure 34. Add-On General Control/Status Register 31 29 2 8 2 7 2 6 2 5 2 4 23 16 15 12 1 1 10 Reserved 0 BIST Condition Code (R/W) nvRAM Operation Address/Data (R/W) Read FIFO Reset (WO) Mailbox Flags Reset (WO) nvRAM Access Failed (RO) nvRAM Access Control AMCC Confidential and Proprietary DS1596 82 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Table 47. Reset General Control/Status Register Bit Description 31:29 nvRAM Access Control. This field provides a method for access to the optional external non-volatile memory. Write operations are achieved by a sequence of byte operations involving these bits and the 8-bit field of bits 23 through 16. The sequence requires that the low-order address, high-order address, and then a data byte are loaded in order. Bit 31 of this field acts as a combined enable and ready for the access to the external memory. D31 must be set to a 1 before an access can begin, and subsequent accesses must wait for bit D31 to become 0 (ready). D31 0 1 1 1 1 0 1 D30 X 0 0 1 1 X X D29 X 0 1 0 1 X X W/R W W W W W R R Inactive Load low address byte Load high address Begin write Begin read Ready Busy Cautionary note: The non-volatile memory interface is also available for access by the Add-On interface. While simultaneous accesses to the nv memory by both the Add-On and PCI are supported, via arbitration logic, software must be designed to prevent the possibility of data corruption within the memory and to provide for accurate data retrieval. 28 27 nvRAM Access Failed. It will indicate that the last nvRAM access has failed. This flag is cleared automatically upon the start of the next read/write operation. Mailbox Flag Reset. Writing a one to this bit causes all mailbox status flags to become reset (EMPTY). It is not necessary to write this bit to 0 afterwards because it is used internally to produce a reset pulse. Since reading this bit will always return a 0, this bit is write only. Reserved. Always zero. Read FIFO Reset. Writing a one to this bit causes the read FIFO to reset (empty). It is not necessary to write a 0 to this bit. This bit is write only. This feature is intended for test only. It can only be asserted when the PCI is not performing any Pass-Thru accesses. Reserved. Always zero. Non-volatile Memory Address/Data Port. This 8-bit field is used in conjunction with bits 31, 30 and 29 of this register to access the external non-volatile memory. The contents written are either low address, high address, or data as defined by bits 30 and 29. This register will contain the external non-volatile memory data when the proper read sequence for bits 31 through 29 is performed. BIST Condition Code. This field is directly connected to the PCI configuration self-test register. Bit 15 through 12 maps with the BIST register bits 3 through 0, respectively. Reserved. Always zero. 26 25 24 23:16 15:12 11:0 AMCC Confidential and Proprietary DS1596 83 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book ADD-ON PASS-THRU CONFIGURATION REGISTER (APTCR) Register Name: Add-On Address: Power-up value: Add-On Attribute: Size: This register controls the configuration for Pass-Thru Regions 1-4. Byte 0 Controls Pass-Thru Region 1 Byte 1 Controls Pass-Thru Region 2 Byte 2 Controls Pass-Thru Region 3 Byte 3 Controls Pass-Thru Region 4 IMPORTANT NOTE: This register (APTCR) is physically the same as the PCI Pass-Thru configuration register (PTCR). It is intended that either the PCI system or local Add-On interface will write to this register, but not both. However, in the event that both the PCI and Add-On write to this register, whichever side wrote last wins. This register is also intended to be initialized once, prior to any PCI bus operations. Pass-Thru operation cannot be guaranteed if this register is updated while a Pass-Thru transaction is being performed. Pass-Thru Configuration Register 60h 80808080h Read/Write 32 bits Figure 35. Pass-Thru Configuration Register 31 Region 4 24 23 Region 3 16 15 Region 2 87 Region 1 0 D7 Pass-Thru Mode 0 - PTADR# Not Driven 1 - PTADR# Driven D6 Endian Conversion 0 - No Conversion 1 - Big Endian Conversion D5 Write FIFO 0 - Enabled 1 - Disabled Active Mode Wait States D2 D1 D0 000 - 0 Wait States 001 - 1 Wait States 010 - 2 Wait States 011 - 3 Wait States 100 - 4 Wait States 101 - 5 Wait States 110 - 6 Wait States 111 - 7 Wait States Prefetch D4 D3 00 - Disabled 01 - Small 10 - Medium 11 - Large AMCC Confidential and Proprietary DS1596 84 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book The following describes one of the four configuration registers. All four region configuration registers are exactly the same. Table 48. Pass-thru Configuration Register Bit Description 7 PTADR# mode. This bit is only valid in Active mode. If this bit is 0, PTADR# is not driven at the beginning of a Active cycle. If this bit is set to 1 (default state), the S5920 will assert PTADR# for one clock cycle after PTATN# is asserted. The Pass-Thru address is also driven while PTADR# is low. This bit is a don’t care if the device is operating in Passive mode Endian conversion. If this bit is set to one, the S5920 will convert the Add-On bus from the default little endian format to a big endian format. Write FIFO disabled. If this bit is set to 1, the S5920 will not accept the next piece of data (on a PCI write) until the Add-On has accepted the previous piece of data. If this bit is set to 0, the S5920 will accept data from the PCI until the Pass-Thru write FIFO is full. Prefetch. These bits control the number of DWORDs that the S5920 will prefetch after the current PCI PassThru read completes. The actual amount of data prefetched depends upon any number of different scenarios. The prefetch values of “small,” “medium” and “large” are available to tune the system to achieve best overall performance (i.e. optimize PCI bus transfers or optimize Add-On bus transfers). The Pass-Thru read FIFO can be enabled to prefetch in either Active mode or Passive mode. Wait states. In Active mode, the user can program the number of wait states required by the Add-On bus to complete a transaction. Up to 7 wait states can be programmed (per region). The S5920 will count the number of clocks programmed into this register before finishing the current data transaction if PTRDY# is high. If PTRDY# is driven low, additional wait states may be inserted. Bits 2, 1 and 0 are don’t care if operating in Passive mode. 6 5 4:3 2:0 AMCC Confidential and Proprietary DS1596 85 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book INTRODUCTION All PCI bus agents and bridges are required to implement PCI Configuration Registers. When multiple PCI devices are present, these registers must be unique to each device in the system. The specified PCI procedure for uniquely selecting a device’s configuration space involves a dedicated signal, called IDSEL, connected to each motherboard PCI bus device and PCI slot. After reset, the host executes configuration cycles to each device on the PCI bus. The configuration registers provide information on PCI agent operation and memory or I/O space requirements. These allow the PCI BIOS to enable the device and locate it within system memory or I/O space. After a PCI reset, the S5920 can be configured for a specific application by downloading device setup information from an external non-volatile memory into the device Configuration Registers. In order to use the Pass-Thru regions, the S5920 must be used with an external nvRAM boot device If no nvRAM is used, the Base-Address Regions are disabled. However, the mailboxes and other PCI/Add-on Operation Registers can still be used (as Base-Address Region #0 comes up in its default state, defining a 128-byte I/O region). To configure the S5920, 64 bytes of setup information are required. The rest of the boot device can be used to implement an expansion BIOS, if desired. Some of the setup information is used to initialize the S5920 PCI Configuration Registers, while other information is used to define S5920 special operating modes. LOADING THE SERIAL NV MEMORY Serial nv memory data transfers are performed through a two-wire, bi-directional data transfer protocol as defined by commercial serial EEPROM offerings. These devices have the advantages of low pin counts, small package size, and economical price. A serial nv memory is initially considered valid if the first serial accesses contain the correct per-byte acknowledgments (see Figure 5). If the serial per-byte acknowledgment is not observed, the S5920 deter mines that no external serial nv memory is present and the AMCC default Configuration Register values specified in the PCI Configuration Register Chapter are used. Please note that the Pass-Thru interface will not operate unless a valid nv memory has been read. The serial nvRAM is first accessed at location 0040h followed by a read to location 0041h. If either of these accesses contain anything other than FFh, the next four accesses are to locations 0050h, 0051h, 0052h and 0053h. At these locations, the data must be 80h (or 81h or 82h), FFh, E8h, and 10h, respectively, for the external nv memory to be considered valid. Once a valid external nv memory has been recognized, it is read, sequentially from location 040h to 07Fh. The data is loaded into the appropriate PCI configuration register. Some of the boot device data is not downloaded into the Configuration Registers, but is used instead to initialize some S5920 modes of operation (location 0045h, for instance). Upon completion of this sequence, the boot load terminates and PCI configuration accesses to the S5920 are acknowledged with the PCI Target Ready (TRDY#) output. Table 1 lists the required nv memory contents for a valid configuration nv memory device. Two pins are used to transfer data between the S5920 PCI controller and the external serial memory: a serial clock pin, SCL, and a serial data pin, SDA. The serial clock pin is an open drain output from the S5920, and the serial data pin is open drain bi-directional. The serial clock is derived by dividing the PCI bus clock by 293. This means the frequency of the serial clock is approximately 114 KHz for a 33 MHz PCI bus clock. Note in Figure 1, a 4.7k pull-up is required on the SDA and SCL lines. During boot-up, the S5920 will only communicate with an EEPROM that has its address pins set to 0 (A[2:0] = “000). When not accessing the external nvRAM, the S5920 will tri-state the SCL and SDA signals so other two-wire serial devices can use the bus. The system designer must guarantee that the two-wire serial bus is idle whenever the S5920 wants to start an access. The S5920 does NOT perform two- PCI RESET Immediately following the assertion of the PCI RST# signal, the Add-On reset output SYSRST# is asserted. The Add-On reset output (SYSRST#) can be used to initialize external state machines, reset Add-On microprocessors, or other Add-On logic devices. All S5920 Operation Registers and Configuration Registers are initialized to their default states at reset. The default values for the Configuration Registers will be overwritten by the contents of the external nv boot memory during device initialization. Configuration accesses by the host CPU while the S5920 is loading configuration will produce PCI bus retries until one of the following events occurs: • The S5920 identifies that there is no valid boot memory (and default Configuration Register values are used). The S5920 finishes downloading all configuration information from a valid boot memory. • AMCC Confidential and Proprietary DS1596 86 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book wire serial arbitration. It assumes that it is the only master on the bus. Communications with the serial memory involve several clock transitions. A start event signals the beginning of a transaction and is immediately followed by an address transfer. Each address/data transfer consists of 8 bits of information followed by a 1-bit acknowledgment. When the exchange is complete, a stop event is issued. Figure 2 shows the unique relationship defining both a start and stop event. Figure 3 shows the required timing for address/data with respect to the serial clock. For random accesses, the sequence involves one clock to define the start of the sequence, eight clocks to send the slave address and read/write command, followed by a one-clock acknowledge, and so on. Figure 4 shows the sequence for a random write access requiring 29 serial clock transitions. At the clock speed Table 49. Valid External Boot Memory Contents Address Data Notes of the S5920, this corresponds to one byte of data transferred approximately every 0.25 milliseconds. Read accesses can be either random or sequential. During boot-up, all accesses from address 40h to 7Fh are sequential. As a result, it is important the nvRAM used supports the nvRAM sequential read accesses as indicated in Figure 6. Figure 5 shows the sequence for a random byte read. To initialize the S5920 controller’s PCI Configuration Registers, the smallest serial device necessary is a 128 x 8 organization. Although the S5920 controller only requires 64 bytes, these configuration bytes must begin at the 64-byte address offset (40h through 7Fh). This offset constraint permits the configuration image to be shared with a memory containing expansion BIOS code and the necessary preamble to identify an expansion BIOS. The largest serial device which can be used is 2 Kbytes. 0040h-0041h 0050h not FFFFh 82h - registers to I/O 81h - memory space 80h - memory below 1 Mbytes FFh E8h 10h This is the location that the S5920 will load a customized vendor ID. (FFFFh is an illegal vendor ID.) This is the least significant byte of the region which initializes the S5920 configuration register BADR0. A value of 81h assigns the 32 DWORD locations of the PCI operations registers into I/O space, a value of 80h defines memory space, and a value of 82h defines memory space below 1 Mbytes. Required Required Required 0051h 0052h 0053h AMCC Confidential and Proprietary DS1596 87 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Figure 36. S5920 to nvRAM Interface Vcc 4.7K S5920 SCL Vcc 4.7K SDA A0 Serial A1 nvRAM A2 24C02 24C04 24C08 24C16 256x8 512x8 1Kx8 2K8 Figure 37. Serial Interface Definition of Start and Stop SCL SDA START BIT STOP BIT Figure 38. Serial Interface Clock Data Relationship SCL SDA DATA STABLE DATA CHANGE DATA STABLE AMCC Confidential and Proprietary DS1596 88 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Figure 39. Serial Interface Byte Access-Write S T A R T R/W SLAVE ADDRESS WORD ADDRESS DATA S T O P A C K 1010 0A C K A C K Figure 40. Serial Interface Byte Access-Read S T A R T R/W SLAVE ADDRESS WORD ADDRESS S T A R T A C K R/W SLAVE ADDRESS DATA S T O P A C K 1010 0A C K 1 010 1A C K Figure 41. Serial Byte Access- Sequential Read S T O P Slave Address Word Address Slave Address Data Data ....... Data AMCC Confidential and Proprietary DS1596 89 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book NON-VOLATILE MEMORY INTERFACE The nv memory, can be accessed through the PCI interface or the Add-On interface. Accesses to the nv memory from the PCI interface are through the Reset Control Register (RCR). Accesses to the nv memory from the Add-On interface are through the Add-On Reset Control Register (ARCR). Some nv memories can contain Expansion ROM BIOS code for use by the host CPU. During initialization, the Expansion BIOS is located within system memory. The starting location of the nv memory is stored in the Expansion ROM Base Address Register in the S5920 PCI Configuration Registers. A PCI read from this region results in the S5920 performing four consecutive byte-wide access to the nv memory device, thus assembling a complete DWORD. Writes to the nv memory are not allowed through the expansion ROM base address region. Any attempt to do so will result in data being accepted by the S5920, but simply discarded. In the RCR and ARCR registers, bits D31:29 are command/status bits and bits D23:16 are address/data bits. These operation registers occupy the same offset (3Ch-3Fh) on their respective interfaces (Add-On or PCI). The sequence used to access the nv memory is the same in either case. the PCI address. For example, if the EXROM Base Address is programmed with 100000h, and the PCI performs a read to address 100040h, this will initiate a read from address 40h of the nvRAM. Once addresses 40h, 41h, 42h and 43h have been read and stored in the nvRAM interface, the S5920 is ready to provide the data to the original PCI device requesting the data. Once the original master comes back to read the data (which it should, as it received a Retry to its initial read), it will get a TRDY# along with the 4 bytes of data that were read from the nvRAM. If the master comes back to retry the read, but the nvRAM interface is not finished with its accesses, the master will again be greeted with a Retry. If a master attempts to read from a different EXROM address, it will also be greeted with a Retry. Only a read with the original address (in our example, a read to address 100040h) will allow the transaction to complete. As a result, if the original master never comes back to Retry the read, the EXROM interface will be hung. Only other EXROM accesses will be hung, as the nvRAM interface will still be operational via the PCI’s RCR and the Add-On’s ARCR. Accesses to the nvRAM via the PCI’s Reset Control Register (RCR) are a bit more involved for the programmer. There are 12 bits of this register that perform both reads and writes. Bits 23-16 to provide Address/ Data information, bits 31-29 are used to provide control information, and bit 28 indicates whether the nvRAM access was successful or not. The control bits 31-29 are assigned as follows (where W/R indicates the type of PCI access to the RCR): D31 0 1 1 1 1 0 1 D30 X 0 0 1 1 X X D29 X 0 1 0 1 X X W/R W W W W W R R nvRAM Interface Function Inactive Load low address byte Load high address byte Begin write Begin read Ready Busy nvRAM READ/WRITE DESCRIPTION There are four different mechanisms to access the external nvRAM: 1. During boot-up (RST# deasserted), the S5920 will automatically read out the nvRAM addresses 40h - 7Fh. 2. Via the PCI Configuration Expansion ROM Base Address Register (EXROM). This is READ-ONLY. 3. Via the PCI Reset Control Register (RCR). This is READ/WRITE. 4. Via the Add-On Reset Control Register (ARCR). This is READ/WRITE. The boot-up sequence is a built-in function, and is affected by the contents of the nvRAM. The Expansion ROM Base Address Register is used if expansion BIOS is stored in the external nvRAM. This register can be enabled for a 2K memory size, and is mapped to access the contents of the nvRAM. When a read is performed to an address in the range of the EXROM base address, a read sequence is started to the nvRAM. As this sequence is extremely slow, the PCI will be greeted with a Retry. Mean-while, the nvRAM interface circuitry will be performing four sequential byte accesses to the nvRAM at the offset indicated by These control bits are used along with the Address/ Data bits 23-16 to configure the type of nvRAM operation (read or write), the address being accessed, and a place to store the write data or the data read from the nvRAM. One can interface with this register in either byte-wide or word-wide fashion. For a word-wide access, the command (bits 31-29) and Address/Data (bits 23-16) are written to the RCR with one PCI write. For a byte-wide access, the command (bits 31-29) is written first, followed by the Address/Data (bits 23-16). This takes two PCI transfers. AMCC Confidential and Proprietary DS1596 90 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book When performing a byte-wide RCR access, users need to write the command indicating how the data is to be used, followed by the data. These commands will assert the internal signals LOAD_LOW_ADDR, LOAD_HIGH_ADDR or LOAD_WR_DATA. Only one signal is asserted at any time: once one is asserted, the others are deasserted. The final read/write interface to the external nvRAM is via the Add-On Reset and Control Register (ARCR). This mechanism is identical to that used for the PCI’s RCR, except that the Add-On interface is used to access the nvRAM via the ARCR. The latency is a bit longer as well, due to the synchronization that must be performed between the Add-On clock and the PCI clock. While on-chip arbitration logic allows simultaneous accesses to the nvRAM via the PCI’s RCR and AddOn’s ARCR (by queuing up the commands), there is no logic to prevent each interface from overwriting nvRAM contents. If an interface writes to a memory location that the other interface has already has written to, the value at that location will be overwritten. What follows are the sequence of steps required to access the nvRAM via the RCR. All the scenarios assume that the RCR is being controlled via PCI bus transactions. By replacing RCR with ARCR in the examples below, the operations are identical for an Add-On device. The following sequence is used to perform nvRAM writes when accessing the RCR/ARCR in a bytewide fashion: 5. Write to RCR(23:16) with the high address byte. Since signal LOAD_HIGH_ADDR is asserted, the data will be written to register NVRAM_HIGH_ADDR. Note that as the nvRAM address is limited to 11 bits, only the 3 lsbs of this write data is actually used. As long as LOAD_HIGH_ADDR is asserted, a write to RCR(23:16) will continue to overwrite register NVRAM_HIGH_ADDR. 6. Write to RCR(31:29) = “000”, a dummy command to deassert either LOAD_LOW_ADDR or LOAD_HIGH_ADDR (whichever occurred last), and to assert internal signal LOAD_WR_DATA. This signal is used to enable the loading of the write data register. LOAD_WR_DATA will remain asserted until another command is issued (load low/high address, begin read/write). As long as L O A D _ W R _ D ATA i s a s s e r t e d , a w r i t e t o RCR(23:16) will continue to overwrite the write data register. 7. Write to RCR(23:16) the byte to be written. Since the signal LOAD_WR_DATA is asserted, the data will be written to the write data register. 8. Write to RCR(31:29) = “110”, the command to start the nvRAM write operation. This will lead to the deassertion of LOAD_WR_DATA and will set the busy bit, RCR(31). The nvRAM interface controller will now initiate a write operation with the external nvRAM. 9. Poll the busy bit until it is no longer set. Once cleared, it is now safe to perform another write/ read operation to the external nvRAM. The XFER_FAIL flag (bit 28) can be used to determine whether the transfer was successful or not. If XFER_FAIL is asserted, this indicates that a transfer to the nvRAM did not receive an ACKNOWLEDGE, and the write transfer should not be considered successful. This flag remains set until the start of the next read/write operation. The busy bit will remain set until the nvRAM interface has completed writing the data byte to the external nvRAM, and has verified that the write sequence is finished. The nvRAM “shuts down” during a write and will not accept any new commands (does not generate an ACKNOWLEDGE) until it finishes the write operation. The S5920 will continue to send commands to the nvRAM until it responds with an ACKNOWLEDGE, after which it clears the busy bit, indicating that the write operation is truly complete. If the busy bit were to be cleared after the nvRAM interface finished the write, but before the external nvRAM was actually finished, a scenario exists where a successive write would be ignored. In this case, the software driver DS1596 91 1. Verify that busy bit, RCR(31), is not set by reading RCR(31). If set, hold off starting the write sequence (repeat step 1 until this bit clears). 2. Write to RCR(31:29) = “100”, the command to load the low address byte. This will assert the internal signal LOAD_LOW_ADDR, which is used to enable the loading of the low-address register (NVRAM_LOW_ADDR). 3. Write to RCR(23:16) with the low address byte. Since signal LOAD_LOW_ADDR is asserted, the data will be written to register NVRAM_LOW_ADDR. As long as LOAD_LOW_ADDR is asserted, a write to RCR(23:16) will continue to overwrite register NVRAM_LOW_ADDR. 4. Write to RCR(31:29) = “101”, the command to load the high address byte. This will assert the internal signal LOAD_HIGH_ADDR, which is used to enable the loading of the high-address register (NVRAM_HIGH_ADDR). AMCC Confidential and Proprietary S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book could not use the busy bit to determine when to start a new write, but would need to insert a delay (deter mined by the “shut down” time of the nvRAM, between 5-10 ms). Fortunately, the S5920 implements the Acknowledge Polling scheme described above, which will not take away the busy bit until the write is truly finished, and the external nvRAM is available for accesses. The following sequence is used to perform nvRAM writes when accessing the RCR/ARCR in a bytewide fashion: the read data in RCR(32:16) may not be valid. This flag remains set until the start of the next read/write operation. W hen performing a word/double-word RCR access, you can combine the data and control in the same command. The following is the sequence for a write: 1. Verify that busy bit, RCR(31), is not set by reading RCR(31). If set, hold off starting the write sequence (repeat step 1 until the bit clears). 2. Write to RCR(31:29) = “100” and RCR(23:16) with the low address byte. This will directly load NVRAM_LOW_ADDR with RCR(23:16). 3. Write to RCR(31:29) = “101” and RCR(23:16) with the high address byte. This will directly load NVRAM_HIGH_ADDR with RCR(23:16). 4. Write to RCR(31:29) = “110” and RCR(23:16) with the write data. This will directly load the write data register with RCR(23:16). This will also set the busy bit, RCR(31). The nvRAM interface controller will now initiate a write operation to the external nvRAM. 5. Poll the busy bit until it is no longer set. Once cleared, it is now safe to perform another write/ read operation to the external nvRAM. In addition, evaluate the XFER_FAIL flag (bit 28) to determine whether the transfer was successful or not. If XFER_FAIL is asserted, this indicates that a transfer to the nvRAM did not receive an ACKNOWLEDGE. The write should not be considered successful. This flag remains set until the start of the next read/write operation. The following sequence is used for a read: 1. Verify that busy bit, RCR(31), is not set by reading RCR(31). If set, hold off starting the read sequence (repeat step 1 until this bit clears). 2. Write to RCR(31:29) = “100”, the command to load the low address byte. This will assert the internal signal LOAD_LOW_ADDR, which is used to enable the loading of the low-address register (NVRAM_LOW_ADDR). 3. Write to RCR(23:16) with the low address byte. Since signal LOAD_LOW_ADDR is asserted, the data will be written to the register NVRAM_LOW_ADDR. As long as LOAD_LOW_ADDR is asserted, a write to RCR(23:16) will continue to overwrite register NVRAM_LOW_ADDR. 4. Write to RCR(31:29) = “101”, the command to load the high address byte. This will assert the internal signal LOAD_HIGH_ADDR, which is used to enable the loading of the high-address register (NVRAM_HIGH_ADDR). 5. Write to RCR(23:16) with the high address byte. Since signal LOAD_HIGH_ADDR is asserted, the data will be written to the register NVRAM_HIGH_ADDR. Note that as the nvRAM address is limited to 11-bits, only the 3-lsb’s of this write data is actually used. As long as LOAD_HIGH_ADDR is asserted, a write to RCR(23:16) will continue to overwrite register NVRAM_HIGH_ADDR. 6. Write to RCR(31:29) = “111”, the command to start the nvRAM read operation. This will set the busy bit, RCR(31). The nvRAM interface controller will now initiate a read operation to the external nvRAM. 7. Poll the busy bit until it is no longer set. Once c l e a r e d , t h e r e a d d a ta w i l l b e l o c a t e d i n RCR(23:16). In addition, evaluate the XFER_FAIL flag (bit 28) to determine whether the transfer was successful or not. If XFER_FAIL is asserted, this indicates that a transfer to the nvRAM did not receive an ACKNOWLEDGE, and AMCC Confidential and Proprietary 1. Verify that the busy bit, RCR(31), is not set by reading RCR(31). If set, hold off starting the read sequence (repeat step 1 until the bit clears). 2. Write to RCR(31:29) = “100” and RCR(23:16) with the low address byte. This will directly load NVRAM_LOW_ADDR with RCR(23:16). 3. Write to RCR(31:29) = “101” and RCR(23:16) with the high address byte. This will directly load NVRAM_HIGH_ADDR with RCR(23:16). 4. Write to RCR(31:29) = “111”. This will set the busy bit, RCR(31). The nvRAM interface controller will now initiate a read operation with the external nvRAM. 5. Poll the busy bit until it is no longer set. Once c l e a r e d , t h e r e a d d a ta w i l l b e l o c a t e d i n RCR(23:16). In addition, evaluate the XFER_FAIL flag (bit 28) to determine whether the DS1596 92 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book transfer was successful or not. If XFER_FAIL is asserted, this indicates that a transfer to the nvRAM did not receive an ACKNOWLEDGE. The read data in RCR(32:16) should not be considered valid. This flag remains set until the start of the next read/write operation. access is a Type 0 configuration cycle or a Type 1 configuration cycle. Type 0 cycles have AD0 and AD1 equal to 0 and are used to access PCI bus agents. Type 1 configuration cycles are intended only for bridge devices and have AD0 as a 1 with AD1 as a 0 during the address phase. The S5920 PCI device is a bus agent (not a bridge) and responds only to a Type 0 configuration accesses. Figure 7 depicts the state of the AD bus during the address phase of a Type 0 configuration access. The S5920 controller does not support the multiple function numbers field (AD[10:8]) and only responds to the all-0 function number value. PCI BUS CONFIGURATION CYCLES Cycles beginning with the assertion IDSEL and FRAME# along with the two configuration command states for C/BE[3:0] (configuration read or write) access the selected device’s configuration space. During the address phase of the configuration cycle just described, the values of AD0 and AD1 identify if the Figure 42. PCI AD Bus Definition Type 0 Configuration Access 31 00h RESERVED 11 10 000 87 REGISTER NUMBER 210 00 TYPE 0 FUNCTION NUMBER XXXXXXXX - INTERNAL REGISTER ADDRESS (DEVICE ID, ETC.) The configuration registers for the S5920 PCI controller can only be accessed under the following conditions: • IDSEL high (PCI slot unique signal which identifies access to configuration registers) along with FRAME# low. Address bits A0 and A1 are 0 (Identifies a Type 0 configuration access). Address bits A8, A9, and A10 are 0 (Function number field of 0 supported). Command bits, C/BE[3:0]# must identify a configuration cycle command (101X). Figure 43. Type 0 Configuration Read Cycles 0 PCI CLK FRAME# AD[31:0] C/BE[3:0]# IRDY# TRDY# IDSEL DEVSEL# ADD 1010 DATA BYTE EN 12345 • • • Figure 8 describes the signal timing relationships for configuration read cycles. Figure 9 describes configuration write cycles. AMCC Confidential and Proprietary DS1596 93 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Figure 44. Type 0 Configuration Write Cycles 0 PCI CLK FRAME# AD[31:0] C/BE[3:0]# IRDY# TRDY# IDSEL DEVSEL# ADD 1011 DATA BYTE EN 12345 EXPANSION BIOS ROMS This section provides an example of a typical PC-compatible expansion BIOS ROM. Address offsets 040h through 07Fh represent the portion of the external nv memory used to boot-load the S5920 controller. Whether the expansion ROM is intended to be executable code is determined by the contents of the first three locations (starting at offset 0h) and a byte checksum over the defined length. The defined length is specified in the byte at address offset 0002h. Table 2 lists each field location by its address offset, its length, its value, and description. The following represents the boot-load image for the S5920 controller’s PCI configuration register: Table 50. PC Compatible Expansion ROM Byte Offset Byte Length (decimal) Binary Value Description Example 0h 1h 2h 3h 7h-17h 18h-19h 1Ah-3Fh 40h 42h 44h 45h 46h 48h 49h 4Ch 4Dh 4Eh 4Fh 1 1 1 4 17 2 38 2 2 1 1 2 1 3 1 1 1 1 55h AAh variable variable variable variable variable Vendor ID Device ID not used S5920 Special Modes not used Revision ID class code not used not used your header type self-test, if desired BIOS ROM signature byte 1 BIOS ROM signature byte 2 Length in multiples of 512 bytes Entry point for INIT function. Reserved (application unique data) Pointer to PCI Data Structure user-defined (see page 2-23) (see page 2-24) 55h AAh 01h 10E8h 5920h xxh (see page 2-89 and 2-129) 01h xxxxh (see page 2-29) (see page 2-30) 00h FF0000h xxh xxh (see page 2-37) (see page 2-38) 00h 80h or 00h AMCC Confidential and Proprietary DS1596 94 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Table 50. PC Compatible Expansion ROM (Continued) Byte Offset Byte Length (decimal) Binary Value Description Example 50h 51h 52h 53h 54h 58h 5Ch 60h 64h 68h 6Ch 6Eh 70h 1 1 1 1 4 4 4 4 4 8 2 2 4 80h, 81h or 82h FFh E8h 10h base addr. #1 base addr. #2 base addr. #3 base addr. #4 not used not used SVID SID (required, see page 2-39, and page 2-74, Table 1) (required per Table 1) (required per Table 1) (required per Table 1) (see page 2-39) (see page 2-39) (see page 2-39) (see page 2-39) 80h, 81h or 82h FFh E8h 10h 00000000h 00000000h 00000000h 00000000h 00000000h XXh (see page 2-44) (see page 2-45) [Expansion ROM base addr.] 5555h 3333h (see page 2-46) (example shows 2K bytes) 74h 7Ch 7Dh 7Eh 7Fh 80h Ñ 1FFh, 2FFh 3FFh etc. 8 1 1 1 1 not used Interrupt line Interrupt pin not used not used FFFFF801h XXh (see page 2-48) (see page 2-49) (see page 2-48) (see page 2-49) application specific 0Ch 01h xxh xxh Byte checksum, location dependent on value for length field at offset 0002h. AMCC Confidential and Proprietary DS1596 95 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book A 16-bit pointer at location 18h of the PC expansion ROM identifies the start offset of the PCI data structure. The PCI data structure is shown in Table 3 and contains various vendor, product, and program descriptions. This structure is provided here for reference only - the user should refer to the PCI BIOS specification for complete details. Note: The access time for large serial devices should be considered, since it may cause a lengthy system delay during initialization. For example, a 2 Kbytes serial device will take about 1 second to be read. Many systems, even when BIOS ROMs are ultimately shadowed into system RAM, may read this memory space twice (once to validate its size and checksum, and once to move it into RAM). Table 51. PCI Data Structure Byte Offset Byte Length (decimal) Binary Value Description 0h 4h 6h 8h Ah Ch Dh 10h 12h 14h 15h 16h 4 2 2 2 2 1 3 2 2 2 1 2 ÔPCIRÕ variable variable variable variable variable variable variable variable variable variable 0000h Signature, the ASCII string ÔPCIRÕ where ÔPÕ is at offset 0, ÔCÕ at offset 1, and so on Vendor Identification Device Identification Pointer to Vital Product Data PCI Data Structure Length (starts with signature field) PCI Data Structure Revision (=0 for this definition) Class Code Image Length Revision Level Code Type Indicator (bit D7=1 signifies “last image”) Reserved AMCC Confidential and Proprietary DS1596 96 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book PCI BUS INTERFACE This section details various events which may occur on the S5920 PCI bus interface. Since the S5920 functions as a target or slave device, signal timing details are given for target transactions only. tiator-target pair. A data phase consists of at least one PCI clock. FRAME# is deasserted to indicate that the final data phase of a PCI cycle is occurring. Wait states may be added to any data phase (each wait state is one PCI clock). The PCI bus command presented on the C/BE[3:0]# pins during the address phase can represent 16 possible states. Table 1 lists the PCI commands and those which are supported by the S5920. A “Yes” in the “Supported by S5920” column in Table 1 indicates that the S5920 device will assert the signal DEVSEL# when that particular command is issued along with the appropriate PCI address. The completion or termination of a PCI cycle can be signaled in several ways. In most cases, the completion of the final data phase is indicated by the assertion of the ready signals from both the target (TRDY#) and initiator (IRDY#) while FRAME# is inactive. In some cases, the target is not able to continue or support a burst transfer and will assert a STOP# signal. This is referred to as a target disconnect. There is also the case where an addressed device does not exist, and the signal DEVSEL# is not driven. In this case, the initiator is responsible for ending the cycle. This is referred to as a master abort. The bus is returned to the idle phase when both FRAME# and IRDY# are deasserted. PCI BUS TRANSACTIONS Because the PCI bus utilizes multiplexed address/data pins (AD[31:0]), every PCI bus transaction consists of an address phase followed by a data phase. An address phase is defined as the clock period in which FRAME# transitions from inactive to active. During the address phase, a bus command is driven by the initiator on the C/BE[3:0]# signal pins. If the command indicates a PCI read, the clock cycle following the address phase is used to perform a “bus turn-around” cycle. A turn-around cycle is a clock period in which the address/data bus is not driven by an initiator or a target device. This is used to avoid PCI bus contention. For a write command, a turn-around cycle is not needed, and the bus goes directly from an address phase to a data phase. All PCI bus transactions consist of an address phase followed by one or more data phases. During the onePCI-clock-long address phase, the bus address and command information is latched into the S5920. The number of data phases depends on how many data transfers are desired or are possible within a given iniTable 52. PCI Bus Commands C/BE[3:0]# Command Type Supported 0000 0001 0010 0011 0100 0101 0110 0111 1000 1001 1010 1011 1100 Interrupt Acknowledge Special Cycle I/O Read I/O Write Reserved Reserved Memory Read Memory Write Reserved Reserved Configuration Read Configuration Write Memory Read Multiple No No Yes Yes No No Yes Yes No No Yes Yes Yes 1 AMCC Confidential and Proprietary DS1596 97 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Table 52. PCI Bus Commands (Continued) C/BE[3:0]# Command Type Supported 1101 1110 1111 Reserved Memory Read Line Memory Write and Invalidate No Yes 1 Yes 2 1. Memory Read Multiple and Memory Read Line are executed as a Memory Read. 2. Memory Write and Invalidate is executed as a Memory Write. PCI BURST TRANSFERS The PCI bus, by default, expects burst transfers to be executed. To successfully perform a burst transfer, both the initiator and target must order their burst address sequence in an identical fashion. There are two different ordering schemes: linear address incrementing and 80486 cache line fill sequencing. The S5920 supports only linear burst ordering. Attempts to perform burst transfers with a scheme other than this will cause the STOP# signal to be asserted during the first data phase, thus issuing a disconnect to the initiator. The S5920 completes the initial data phase successfully, but asserting STOP# indicates that the next access needs to be a completely new cycle. Some accesses to the S5920 controller do not support burst transfers. For example, the S5920 does not allow burst transfers when accesses are made to the configuration or operation registers. Attempts to perform burst transfers to these regions will cause a disconnect on the PCI bus, as described above. Expansion ROM accesses also do not support bursts, and will respond in the same way. Accesses to memory or I/O regions defined by the Base Address Registers 1-4 may be bursts, if desired. asserted, the initiator is just adding wait states, not necessarily attempting a burst. Figure 45. Single Data Phase PCI Bus Read of S5920 Registers or Expansion ROM 12345 PCLK FRAME# (I) AD[31:0] C/BE[3:0]# IRDY# TRDY# DEVSEL# STOP# (I) Driven by Initiator (T) Driven by Target Address Bus Cmd (T) Data Byte Enables PCI READ TRANSFERS The S5920 responds to PCI bus memory or I/O read transfers when it is selected as a target. PCI targets may drive DEVSEL# and TRDY# after the end of the address phase. TRDY# is not driven until the target can provide valid data for the PCI read. Read accesses from the S5920 operation registers are shown in Figure 1. The S5920 conditionally asserts STOP# in clock period 3 if the initiator keeps FRAME# asserted during clock period 2 with IRDY# asserted (indicating a burst is being attempted). Wait states may be added by the initiator by not asserting the signal IRDY# during clock 3 and beyond. If FRAME# remains asserted, but IRDY# is not There are only two conditions where accesses to the S5920 do not return TRDY#, but assert STOP# instead. This condition is called a target-initiated termination or target disconnect. This can occur when a read attempt is made to an empty Pass-Thru FIFO. The second condition may occur when read accesses to the expansion ROM generate a retry if the nvRAM interface has not finished reading 4 bytes. When burst read transfers are attempted to the S5920 operation registers, configuration registers or expansion ROM, STOP# is asserted during the first data transfer to indicate to the initiator that no further transfers (data phases) are possible. This is a targetinitiated termination where the target disconnects after the first data phase. Figure 2 shows the signal relationships during a burst read attempt to the S5920 operation registers. DS1596 98 AMCC Confidential and Proprietary S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book PCI WRITE TRANSFERS Write transfers on the PCI bus are one clock period shorter than read transfers. This is because the AD[31:0] bus does not require a turn-around cycle between the address and data phases. Write accesses to the S5920 operation registers are shown in Figure 3. Here, the S5920 asserts the signal STOP# in clock period 3. STOP# is asserted because the S5920 does not support burst writes to operation registers. Wait states may be added by the initiator by not asserting the signal IRDY# during clock 2 and beyond. There is only one condition where writes to S5920 internal registers do not return TRDY# (but do assert STOP#). This is called a target-initiated termination or target disconnect. This occurs when a write attempt is made to a full Pass-Thru FIFO. The assertion of STOP# without the assertion of TRDY# indicates that the initiator should retry the operation later. The S5920 will sustain a burst as long as the FIFO is not full. Figure 46. Burst PCI Bus Read Attempt to S5920 Registers or Expansion ROM 12345 PCLK FRAME# (I) AD[31:0] C/BE[3:0]# IRDY# TRDY# DEVSEL# STOP# (I) Driven by Initiator (T) Driven by Target Address Bus Cmd (T) Data Byte Enables Figure 47. Burst PCI Bus Write of S5920 Registers 1 PCLK FRAME# AD[31:0] (I) C/BE[3:0]# IRDY# TRDY# DEVSEL# STOP# (I) (I) (T) (T) (T) Data Transfered Bus Cmd BE 1 BE 2 (I) Address Data 1 Data 2 2 3 4 5 (I) Driven by Initiator (T) Driven by Target No Data Transfered Target-Initiated Termination There are situations where the target may end a transfer prematurely. This is called “target-initiated termination.” Target termination falls into three categories: disconnect, retry, and target abort. Only the disconnect termination completes a data transfer. Target Disconnects There are many situations where a target may disconnect. Slow responding targets may disconnect to permit more efficient (faster) devices to be accessed while they prepare for the next data phase. Or a target may disconnect if it recognizes that the next data phase in a burst transfer is out of its address range. A target disconnects by asserting STOP#, TRDY#, and DEVSEL# as shown in Figures 4a and 4b. The initiator in Figure 4a responds to the disconnect condition by deasserting FRAME# on the following clock but does not complete the data transfer until IRDY# is asserted. The timing diagram in Figure 4b also applies to the S5920. The S5920 performs a target disconnect if a burst access is attempted to any of its PCI Operation/Configuration Registers, or to the Expansion ROM. AMCC Confidential and Proprietary DS1596 99 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Target Requested Retries Figure 49. Figure 4b. Target Disconnect Example 2 1 PCLK FRAME# IRDY# TRDY# DEVSEL# ST OP# (I) (I) (T) (T) (T) Target Disconnect Single Data Tr ansferr ed (I) Driven by Initiato (T) Driven by Targ e 2 3 4 5 The S5920 initiates a retry for Pass-Thru writes when the Write FIFO is full, and for Pass-Thru reads when the Add-On cannot supply data within 16 PCI clocks from the assertion of FRAME# (for the first data phase of a burst). A retry is requested by a target by asserting both STOP# and DEVSEL# while TRDY# is deasserted. Figure 5 shows the behavior of the S5920 when performing a target-initiated retry. Figure 48. Figure 4a. Target Disconnect Example 1 1 PCLK FRAME# IRDY# TRDY# DEVSEL# STOP# (I) (I) (T) (T) (T) Target Disconnect Identified (I) Driven by Initiator (T) Driven by Target 2 3 4 5 Data Transfered Target Aborts Data Transfered A target abort termination represents an error condition when no number of retries will produce a successful target access. A target abort is uniquely identified by the target deasserting DEVSEL# and TRDY# while STOP# is asserted. When a target performs an abort, it must also set bit 11 of its PCI Status register (PCISTS). The S5920 never responds with a target abort when accessed. Target termination types are summarized in Table 2. AMCC Confidential and Proprietary DS1596 100 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Target Latency Figure 50. Figure 5. Target-Initiated Retry 1 PCLK FRAME# IRDY# TRDY# DEVSEL# STOP# (I) (I) (T) (T) (T) Initiator Sequences IRDY# + FRAME# to return to IDLE state 2 3 4 5 The PCI specification requires that a selected target relinquish the bus should an access to that target require more than eight PCI clock periods (16 clocks for the first data phase, 8 clocks for each subsequent data phase). This prevents slow target devices from potentially monopolizing the PCI bus and also allows more accurate estimations for bus access latency. Note that a special mode is available to the user which will allow for this mechanism to be disabled, thus violating the PCI 2.1 Specification. If a value of 0 is programmed into the serial nvRAM location 45h, bit 0, target latency is ignored. In this case, the S5920 will never issue a retry/disconnect in the event of a slow Add-On device. This programmable bit is only pro vided for flexibility, and most users should leave this bit set to 1. nvRAM Location 45h, bit 0 = 0 : No disconnect for slow Add-On device. nvRAM Location 45h, bit 0 = 1 : PCI 2.1 compliant Target Locking (I) Driven by Initiator Target Retry (T) Driven by Target Signaled Figure 51. Figure 6. Engaging the LOCK# Signal 1 PCLK FRAME# LOCK# (I) AD[31:0] Address (T ) Data (I) 2 3 4 5 It is possible for a PCI bus master to obtain exclusive access to a target (“locking”) through use of the PCI bus signal LOCK#. LOCK# is different from the other PCI bus signals because its ownership may belong to any bus master, even if it does not currently have ownership of the PCI bus. The ownership of LOCK#, if not already claimed by another master, may be achieved by the current PCI bus master on the clock period following the initial assertion of FRAME#. Figure 6 describes the signal relationship for establishing a lock. The ownership of LOCK#, once established, persists even while other bus masters control the bus. Ownership can only be relinquished by the master which originally established the lock. PCI Bus Access Latency Components Bus Access Latency IRDY# TRDY# DEVSEL# (I) (T) (T) LOCK MECHANISM AVAILABLE UPON FIRST ACCESS (I) Driven by Initiator (T) Driven by Target TARGET BECOMES LOCKED LOCK MECHANISM AVAILABLE BUS IDLE LOCK ESTABLISHED LOCK MAINTAINED REQ# Asserted GNT# Asserted FRAME# Asserted TRDY# Asserted --Arbitration Latency-- --Bus Acquisition-Latency --Target Latency-- AMCC Confidential and Proprietary DS1596 101 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Table 53. Target Termination Type Termination DEVSEL# STOP# TRDY# Comment Disconnect Retry Abort on on off on on on on off off Data is transferred. Transaction needs to be re-initiated to complete. Data was not transferred. Transaction should be tried later. Data was not transferred. Fatal error. Targets selected with LOCK# deasserted during the assertion of FRAME# (clock period 1 of Figure 6), which encounter the assertion of LOCK# during the following clock (clock period 2 of Figure 6) are thereafter considered “locked.” A target, once locked, requires that subsequent accesses to it deassert LOCK# while FRAME# is asserted. Figure 7 shows a subsequent access to a locked target by the master which locked it. Because LOCK# is owned by a single master, only that master is able to deassert it at the beginning of a transaction (assuming successful access to the locked target). A locked target can only be unlocked during the clock period following the last data transfer of a transaction when the LOCK# signal is deasserted. An unlocked target ignores LOCK# when it observes that LOCK# is already asserted during the first clock period of a transaction. This allows other masters to access other (unlocked) targets. If an access to a locked target is attempted by a master other than the one that locked it, the target responds with a retry request, as shown in Figure 8. The S5920 responds to and supports bus masters which lock it as a target. AMCC Confidential and Proprietary DS1596 102 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book PCI BUS INTERRUPTS The S5920 controller is able to generate PCI bus interrupts by asserting the PCI bus interrupt signal (INTA#). INTA# is a multi-sourced, wire-ORed signal on the PCI bus and is driven by an open drain output on the S5920. The assertion and deassertion of INTA# have no fixed timing relationship with respect to the PCI bus clock. Once the S5920 asserts INTA#, it remains asserted until the interrupt source is cleared by a write to the Interrupt Control/Status Register (INTCSR). In the case of the external Add-On Interrupt, INTA# will remain set as long as the ADDINT# pin is driven low by an Add-On device(s). The source(s) driving ADDINT# must deassert this input before the PCI interrupt (INTA#) is driven to the false state. It is the responsibility host software to clear the Add-On interrupt source before exiting its interrupt handler routine. The S5920 asserts SERR# if it detects odd parity during an address phase, if enabled. The SERR# enable bit is bit 8 in the S5920 PCI Command Register (PCICMD). The odd parity error condition involves the state of signals AD[31:0] and C/BE[3:0]# when FRAME# is first asserted and the PAR signal during the following clock. If an error is detected, the S5920 asserts SERR# on the following (after PAR valid) clock. Since many targets may observe an error on an address phase, the SERR# signal is an open-drain multi-sourced, wire-ORed signal on the PCI bus. The S5920 drives SERR# low for one clock period when an address phase error is detected. Once an SERR# error is detected by the S5920, the PCI Status register bit 14, System Error, is set and remains set until cleared through software or a hardware reset. The PERR# signal is similar to the SERR# with two differences: it reports errors for the data phase and is only asserted by the device receiving the data. The S5920 drives this signal (removed from tri-state) when it is the selected target for write transactions. The parity error conditions are only reflected by the PERR# pin if the Parity Error Enable bit (bit 6) of the PCI Command Register is set. Upon the detection of a data parity error, the Detected Parity Error bit (bit 15) of the PCI Status Register is set (PCISTS). Unlike the PERR# signal pin, this Status bit is set regardless of the state of the PCI Command Register's Parity Error Enable bit. The assertion of PERR# occurs two clock periods following the data transfer. This two-clock delay occurs because the PAR signal does not become valid until the clock following the transfer, and an additional clock is provided to generate and assert PERR# once an error is detected. PERR# is only asserted for one clock cycle for each error sensed. The S5920 only qualifies the parity error detection during the actual data transfer portion of a data phase (when both IRDY# and TRDY# are asserted). PCI BUS PARITY ERRORS The PCI specification defines two error-reporting signals, PERR# and SERR#. These signals indicate a parity error condition on the signals AD[31:0], C/ BE[3:0]#, and PAR. The validity of the PAR signal is delayed one clock period from its corresponding AD[31:0] and C/BE[3:0]# signals. Even parity is supported by PCI: when the total number of ones in the group of signals AD[31:0] and C/BE[3:0]# is equal to an even number the parity bit will be deasserted. If an odd number of ones is seen, the parity bit will be asserted. PERR# is the error-reporting mechanism for parity errors that occur during the data phase for all but PCI Special Cycle commands. SERR# is the error-reporting mechanism for parity errors that occur during the address phase. The timing diagram in Figure 9 shows the timing relationships between the signals AD[31:0], C/BE[3:0]#, PAR, PERR# and SERR#. AMCC Confidential and Proprietary DS1596 103 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Figure 52. Access to a Locked Target by its Owner 1 PCLK 2 3 4 5 Figure 53. Access Attempt to a Locked Target 1 PCLK 2 3 4 5 FRAME# (I) FRAME# LOCK# (I) LOCK# (I) AD[31:0] Address Data Data AD[31:0] (I) Address Data IRDY# (I) IRDY# TRDY# (I) (T) TRDY# (T) DEVSEL# (T) DEVSEL# (T) LOCKED TARGET IDENT IFIES OWNER CONDITION WHICH UNLOCKS TARGET (I) Driven by Initiator (T) Driven by Target STOP# (T) LOCKED TARGET IDENTIFIES THAT BUS MASTER IS NOT ITS OWNER (I) Driven by Initiator (T) Driven by Target CAUSES TARGET RETRYTERMINATION Figure 54. Error Reporting Signal 1 PCLK FRAME# AD[31:0] C/BE[3:0] (I) (I) (I) (T) Address A CMD A Data A BE's A (I) Address B CMD B Data B BE's B 2 3 4 5 6 7 8 9 PAR SERR# PERR# (T) (I) GOOD ERROR (T) (T) (I) (I) GOOD GOOD ERROR ERROR GOOD ERROR B A READ TRANSACTION WRITETRANSACTION (I) Driv enby Initiator (T) Driv en by Target AMCC Confidential and Proprietary DS1596 104 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book MAILBOX OVERVIEW The S5920 has two 32-bit mailbox registers. These mailboxes are useful for passing command and status information between the Add-On and the PCI bus. The PCI interface has one incoming mailbox (Add-On to PCI) and one outgoing mailbox (PCI to Add-On). The Add-On interface has one incoming mailbox (PCI to Add-On) and one outgoing mailbox (Add-On to PCI). The PCI incoming and Add-On outgoing mailboxes are the same, internally. The Add-On incoming and PCI outgoing mailboxes are also the same, internally. The mailbox status may be monitored in two ways. The PCI and Add-On interfaces each have a mailbox status register to indicate the empty/full status of data bytes within the mailboxes. The mailboxes may also be configured to generate interrupts to the PCI and/or Add-On interface. The outgoing and the incoming mailbox on each interface can be configured to generate interrupts. Figure 55. Figure 1. PCI to Add-On Mailbox Register MAILBOX REGISTER OUTPUT INTERLOCK REGISTER FUNCTIONAL DESCRIPTION Figure 1 shows a block diagram of the PCI to Add-On mailbox registers. Add-On incoming mailbox read accesses pass through an output interlock register. This prevents a PCI bus write to a PCI outgoing mailbox from corrupting data being read by the Add-On. Figure 2 shows a block diagram of the Add-On to PCI mailbox registers. PCI incoming mailbox reads also pass through an interlocking mechanism. This pre vents an Add-On write to an outgoing mailbox from corrupting data being read by the PCI bus. The following sections describe the mailbox flag functionality and the mailbox interrupt capabilities. PCI BUS DQ DQ ADD-ON BUS LOAD ENABLE EN EN PCI CLK ADCLK RD# SELECT# ADR -MB Q D ADCLK VDD DQ PCI CLK R S MAILBOX FULL AMCC Confidential and Proprietary DS1596 105 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Figure 56. Add-On to PCI Mailbox Register OUTPUT INTERLOCK REGISTER PCI INCOMMING MAILBOX MAILBOX REGISTER PCI BUS Q D Q D ADD-ON BUS EN EN EN PCI CLK PCI READ ADCLK VDD ADCLK D R S Q D ADCLK Q MAILBOX FULL WR# SELECT# ADR -MB Mailbox Empty/Full Conditions The PCI and Add-On interfaces each have a mailbox status register. The PCI Mailbox Empty/Full Status (MBEF) and Add-On Mailbox Empty/Full Status (AMBEF) registers indicate the status of all bytes within the mailbox registers. A write to an outgoing mailbox sets the status bits for that mailbox. The byte enables determine which bytes within the mailbox become full (and which status bits are set). An outgoing mailbox for one interface is an incoming mailbox for the other. Therefore, incoming mailbox stat u s b i ts o n o n e i n t e r f a c e a r e i d e n t i c a l t o t h e corresponding outgoing mailbox status bits on the other interface. The following list shows the relationship between the mailbox registers on the PCI and Add-On interfaces. PCI Interface Outgoing Mailbox Incoming Mailbox PCI Mailbox Empty/ Full 13 are set. These bits indicate that outgoing mailbox bytes 0 and 1 are full. Reading the Add-On Mailbox Empty/Full Status Register (AMBEF) shows that bits 12 and 13 in this register are also set, indicating the Add-On incoming mailbox bytes 0 and 1 are full. An Add-On read of the incoming mailbox, bytes 0 and 1, clears the status bits in both the MBEF and AMBEF status registers. The read-only status flags in the MBEF and AMBEF registers are reset when the corresponding byte is read from the incoming mailbox. Alternately, these flags can be globally reset from either the PCI interface or the Add-On interface. The PCI Bus Reset Control Register (RCR) and the Add-On Reset Control Register (ARCR) each have a bit to reset all of the mailbox status flags. Mailbox Interrupts = = = Add-On Interface Incoming Mailbox Outgoing Mailbox Add-On Mailbox Empty/ Full A write to an outgoing mailbox also writes data into the incoming mailbox on the other interface. It also sets the status bits for the outgoing mailbox and the status bits for the incoming mailbox on the other interface. Reading the incoming mailbox clears the corresponding status bit(s) in the Add-On and PCI mailbox status registers (AMBEF and MBEF). For example, a PCI write is performed to the PCI outgoing mailbox, writing bytes 0 and 1 (CBE0# and CBE1# asserted). Reading the PCI Mailbox Empty/ Full Status Register (MBEF) indicates that bits 12 and AMCC Confidential and Proprietary The designer has the option to generate interrupts to the PCI and Add-On interfaces when specific mailbox events occur. The PCI and Add-On interfaces can each define two conditions where interrupts may be generated. An interrupt can be generated when the incoming mailbox becomes full and/or when the outgoing mailbox becomes empty. A specific byte within a specific mailbox is selected to generate the interrupt. The conditions defined to generate interrupts to the PCI interface do not have to be the same as the conditions defined for the Add-On interface. For the incoming mailbox interrupts, when the specified byte becomes full, an interrupt is generated. The interrupt might be used to indicate command or status information has been provided, and must be read. For DS1596 106 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book PCI incoming mailbox interrupts, the S5920 asserts the PCI interrupt, INTA#. For Add-On incoming mailbox interrupts, the S5920 asserts the Add-On interrupt, IRQ#. For the outgoing mailbox interrupts, when the specified byte becomes empty, an interrupt is generated. The interrupt might be used to indicate that the other interface has received the last information sent and more may be written. For PCI outgoing mailbox interrupts, the S5920 asserts the PCI interrupt, INTA#. For Add-On outgoing mailbox interrupts, the S5920 asserts the Add-On interrupt, IRQ#. Add-On Outgoing Mailbox, Byte 3 Access When the MD[7:0] bus is set up for I/O mode and LOAD# is high (deasserted), the MD[7:0] bus is an active output, driving the contents of the PCI outgoing mailbox, byte 3 (OMB[3]). In this case, the MD[7:0] bus will be updated anytime the PCI writes to mailbox OMB[3]. As a result, the MD[7:0] bus will be synchronous to the PCI clock. When LOAD# is driven low, the MD[7:0] bus is tri-stated, allowing external data to be latched into Add-On outgoing mailbox byte 3. This is a similar function that exists for input-only. Figures 3 and 4 show the interaction between the MD[7:0] bus and the LOAD# input pin. Note that a turnaround cycle is utilized when writing data to the mailbox byte in I/O mode. This is to prevent contention on the MD[7:0] drivers. PCI incoming mailbox byte 3 (Add-On outgoing mailbox, byte 3, or AOMB[3]) has been further enhanced by the addition of a separate 8-bit interface (MD[7:0]) on the Add-On side. This interface can be used to write to AOMB[3] instead of/or in addition to the normal method (via an Add-On Operation Register write to AOMB[3]). The MD[7:0] bus can be configured in one of two modes: Input mode or I/O mode. If the configuration pin MDMODE is strapped high, the MD[7:0] bus is set to be in input mode only. If MDMODE is strapped low, the MD[7:0] bus will operate in a bi-directional mode. If the MD[7:0] bus is set up for input-only mode, data will be latched into AOMB[3] when the LOAD# input is sampled low by the Add-On clock. The LOAD# input pin may also be used to generate a PCI interrupt if the appropriate interrupt is enabled in the Interrupt Control/Status Register (INTCSR). These functions are identical to the Add-On device writing to its byte 3 outgoing mailbox via the DQ bus. As a matter of fact, Add-On mailbox byte 3 is accessible by either the external mailbox port or the Add-On interface. Whichever interface writes to it last will determine the data that resides in that register. BUS INTERFACE The mailboxes appear on the Add-On and PCI bus interfaces as two operation registers. One is the outgoing mailbox, and the other is the incoming mailbox. These mailboxes may be used to generate interrupts to each of the interfaces. The following sections describe the Add-On and PCI bus interfaces for the mailbox registers. PCI Bus Interface The mailbox operation registers do not support burst accesses by an initiator. A PCI initiator attempting to burst to the mailbox registers causes the S5920 to respond with a target disconnect with data. PCI writes to a full outgoing mailbox overwrite data currently in that mailbox. PCI reads from an empty incoming mailbox return the data that was previously contained in the mailbox. In this case, the data cannot be guaranteed. It is intended for the user to verify that a mailbox is full before it is read. PCI incoming and outgoing mailbox interrupts are enabled/disabled in the INTCSR. The mailboxes can generate a PCI interrupt (INTA#) under two conditions (individually enabled). For an incoming mailbox full interrupt, INTA# is asserted on the rising edge of the PCI clock after the Add-On mailbox write completes. For an outgoing mailbox empty interrupt, INTA# is asserted on the rising edge of the PCI clock after the Add-On mailbox read completes. INTA# is deasserted one PCI clock cycle after the mailbox interrupt is serviced (by writing a 1 to the proper interrupt source bit). Add-On Bus Interface Signal Pin Add-On Outgoing Mailbox MD0 MD1 MD2 MD3 MD4 MD5 MD6 MD7 Mailbox,Bit 24 Mailbox, Bit 25 Mailbox, Bit 26 Mailbox, Bit 27 Mailbox, Bit 28 Mailbox,Bit 29 Mailbox, Bit 30 Mailbox, Bit 31 The Add-On mailbox interface behaves similarly to the PCI bus interface. Add-On writes to a full outgoing mailbox overwrite data currently in that mailbox. PCI reads from an empty incoming mailbox return the data that was previously contained in the mailbox. DS1596 107 AMCC Confidential and Proprietary S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Add-On incoming and outgoing mailbox interrupts are enabled/disabled in the Add-On Interrupt Control/Status Register (AINT). The mailboxes can generate the Add-On IRQ# interrupt under two conditions (individually enabled). For an incoming mailbox full interrupt, IRQ# is asserted on the rising edge of the Add-On clock after the PCI mailbox write completes. For an outgoing mailbox empty interrupt, IRQ# is asserted on the rising edge of the Add-On clock after the PCI mailbox read completes. IRQ# is deasserted one Add-On clock cycle after the mailbox interrupt is serviced (by writing a 1 to the proper interrupt source bit). 8-Bit and 16-Bit Add-On Interfaces CONFIGURATION The PCI interface and the Add-On interface each have one incoming mailbox (IMB or AIMB) and one outgoing mailbox (OMB or AOMB) along with a single mailbox status register (MBEF or AMBEF). The outgoing mailbox is read/write, the incoming mailbox and the mailbox status registers are read-only. The following sections discuss the registers associated with the mailboxes and accesses required for different modes of mailbox operation. Mailbox Status Some Add-On designs may implement an 8-bit or 16bit bus interface. The mailboxes do not require a 32-bit Add-On interface for all the bytes to be read/written. For 8-bit interfaces, the 8-bit data bus may be externally connected to all 4 bytes of the 32-bit Add-On interface (DQ 31:24, 23:16, 15:8, 7:0 are all con nected). The Add-On device reading or writing the mailbox registers may access all mailbox bytes by cycling through the Add-On byte enable inputs (only one byte enable may be active at a time). A similar solution applies to 16-bit Add-On buses. This solution works for Add-On designs which always use just one bus width (8-bit or 16-bit). If the DQMODE pin is high, indicating a 16-bit Add-On interface, the previous solution may be used to implement an 8-bit interface. The only modification needed is that BE3(= ADR1) must be toggled after the first two accesses to steer the S5920 internal data bus to access the upper 16 bits of the mailboxes. Every byte in each mailbox has a status bit in the Mailbox Empty/Full Status Registers (MBEF and AMBEF). Writing a particular byte into the outgoing mailbox sets the corresponding status bit in both the MBEF and AMBEF registers. A read of a ‘full’ byte in a mailbox clears the status bit. The MBEF and AMBEF are readonly. Status bits cannot be cleared by writes to the status registers. The S5920 allows the mailbox status bits to be reset through software. The PCI Bus Reset Control PCI Operation Register (RCR) and the Add-On Reset Control Add-On Operation Register (ARCR) each have a bit to reset mailbox status. Writing a 1 to Mailbox Flag Reset bit in the RCR or the ARCR register immediately clears all bits in the both the MBEF and AMBEF registers. Writing a 0 has no effect. The Mailbox Flag Reset bit is write-only. The flag bits should be monitored when transferring data through the mailboxes. Checking the mailbox status before performing an operation prevents data from being lost or corrupted. The following sequences are suggested for PCI mailbox operations using status polling (interrupts disabled). Figure 57. Input/Output Mode (MDMODE=0) ADCLK LOAD# MD[0:7] S5920 Driving Turn Around WRITE DATA Turn Around S5920 Driving Figure 58. Input Mode (MDMODE=1) ADCLK LOAD# MD[0:7] WRITE DATA AMCC Confidential and Proprietary DS1596 108 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Reading the PCI Incoming Mailbox: 1. Check Mailbox Status. Read the mailbox status register to determine if any information has been passed from the Add-On interface. MBEF Bits 31:28 If a bit is set, valid data is contained in the corresponding mailbox byte. 2. Read Mailbox. Read the mailbox bytes which MBEF indicates are full. This automatically resets the status bits in the MBEF and AMBEF registers. IMB Bits 31:0 Mailbox data. Writing the PCI Outgoing Mailbox: 1. Check Mailbox Status. Read the mailbox status register to determine if information previously written to the mailbox has been read by the Add-On interface. Writes to full mailbox bytes overwrite data currently in the mailbox (if not already read by the Add-On interface). Repeat until the byte(s) to be written are empty. MBEF Bits 15:12 If a bit is set, valid data is contained in the corresponding mailbox byte and has not been read by the Add-On. 2. Write Mailbox. Write to the outgoing mailbox byte(s). OMB Bits 31:0 Mailbox data. Mailbox operations for the Add-On interface are functionally identical. The following sequences are suggested for Add-On mailbox operations using status polling (interrupts disabled): Reading an Add-On Incoming Mailbox: 1. Check Mailbox Status. Read the mailbox status register to determine if any information has been passed from the PCI interface. AMBEF Bits 15:12 If a bit is set, valid data is contained in the corresponding mailbox byte. 2. Read Mailbox. Read the mailbox bytes which AMBEF indicates are full. This automatically resets the status bits in the AMBEF and MBEF registers. AIMB Bits 31:0 Mailbox data. Writing an Add-On Outgoing Mailbox: 1. Check Mailbox Status. Read the mailbox status register to determine if information previously written to the mailbox has been read by the PCI interface. Writes to full mailbox bytes overwrite data currently in the mailbox (if not already read by the PCI interface). Repeat until the byte(s) to be written are empty. AMBEF Bits 31:28 If a bit is set, valid data is contained in corresponding mailbox byte and has not been read by the PCI bus. 2. Write Mailbox. Write to the outgoing mailbox byte(s). AOMB Bits 31:0 Mailbox data. AMCC Confidential and Proprietary DS1596 109 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Mailbox Interrupts Although polling status is useful in some cases, polling requires continuous actions by the processor. Mailbox interrupt capabilities are provided to avoid much of the processor overhead required by continuously polling status bits. The Add-On and PCI interface can each generate interrupts on the incoming mailbox condition and/or the outgoing mailbox condition. These can be individual enabled/disabled. A specific byte in the incoming mailbox and outgoing mailbox is identified to generate the interrupt(s). The tasks required to setup the mailbox interrupts are as follows: Enabling PCI mailbox interrupts: 1. Enable PCI outgoing mailbox interrupts. A specific byte within the outgoing mailboxes is identified to assert INTA# when read by the Add-On interface. INTCSR INTCSR Bit 4 Bits 1:0 Enable outgoing mailbox interrupts Identify mailbox byte to generate interrupt 2. Enable PCI incoming mailbox interrupts. A specific byte within the incoming mailboxes is identified to assert INTA# when written by the Add-On interface. INTCSR INTCSR Bit 12 Bits 9:8 Enable incoming mailbox interrupts Identify mailbox byte to generate interrupt Enabling Add-On mailbox interrupts: 1. Enable Add-On outgoing mailbox interrupts. A specific byte within the outgoing mailboxes is identified to assert IRQ# when read by the PCI interface. AINT AINT Bit 12 Bits 9:8 Enable outgoing mailbox interrupts Identify mailbox byte to generate interrupt 2. Enable Add-On incoming mailbox interrupts. A specific byte within the incoming mailboxes is identified to assert IRQ# when written by the PCI interface. AINT AINT Bit 4 Bits 1:0 Enable incoming mailbox interrupts Identify mailbox byte to generate interrupt With either the Add-On or PCI interface, these two steps can be performed with a single access to the appropriate register. They are shown separately here for clarity. Once interrupts are enabled, the interrupt service routine must access the mailboxes and clear the interrupt source. A particular application may not require all of the steps shown. For instance, a design may only use the incoming mailbox interrupts and not require support for the outgoing mailbox interrupts. The interrupt service routine tasks are as follows: AMCC Confidential and Proprietary DS1596 110 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Servicing a PCI Mailbox Interrupt (INTA# asserted): 1. Identify the interrupt source(s). Multiple interrupt sources are available on the S5920. The interrupt service routine must verify that a mailbox generated the interrupt (and not some other interrupt source). INTCSR INTCSR INTCSR Bit 23 Bit 17 Bit 16 PCI interrupt asserted PCI incoming mailbox interrupt indicator PCI outgoing mailbox interrupt indicator 2. Check mailbox status. The mailbox status bits indicate which mailbox bytes must be read or written. MBEF MBEF Bits 31:28 Bits 15:12 Full PCI incoming mailbox bytes Empty PCI outgoing mailbox bytes 3. Access the mailbox. Based on the contents of MBEF, mailboxes are read or written. Reading an incoming mailbox byte clears the corresponding status bit in MBEF. OMB IMB Bits 31:0 Bits 31:0 PCI outgoing mailboxes PCI incoming mailboxes 4. Clear the interrupt source. The PCI INTA# signal is deasserted by clearing the interrupt request. The request is cleared by writing a 1 to the appropriate bit. INTCSR INTCSR Bit 17 Bit 16 Clear PCI incoming mailbox interrupt Clear PCI outgoing mailbox interrupt Servicing the Add-On mailbox interrupt (IRQ# asserted): 1. Identify the interrupt source(s). Multiple interrupt sources are available on the S5920. The interrupt service routine must verify that a mailbox generated the interrupt (and not some other interrupt source). AINT AINT AINT Bit 23 Bit 17 Bit 16 Add-On interrupt asserted Add-On outgoing mailbox interrupt indicator Add-On incoming mailbox interrupt indicator 2. Check mailbox status. The mailbox status bits indicate which mailbox bytes must be read or written. AMBEF AMBEF Bits 31:28 Bits 15:12 Empty Add-On outgoing mailbox bytes Full Add-On incoming mailbox bytes 3. Access the mailbox. Based on the contents of AMBEF, mailboxes are read or written. Reading the incoming mailbox byte clears the corresponding status bit in AMBEF. AIMB AOMB Bits 31:0 Bits 31:0 Add-On incoming mailbox Add-On outgoing mailbox AMCC Confidential and Proprietary DS1596 111 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book 4. Clear the interrupt source. The Add-On IRQ# signal is deasserted by clearing the interrupt request. The request is cleared by writing a 1 to the appropriate bit. AINT AINT Bit 17 Bit 16 Clear Add-On outgoing mailbox interrupt Clear Add-On incoming mailbox interrupt NOTE: For an incoming mailbox interrupt, step 3 involves accessing the mailbox. To allow the incoming mailbox interrupt logic to be cleared, the mailbox status bit must also be cleared. Reading an incoming mailbox clears the status bits. Another option for clearing the status bits is to use the Mailbox Flag Reset bit in the RCR and ARCR registers, but this clears all status bits, not just a single mailbox byte. For outgoing mailbox interrupts, the status bit was already cleared prior to the generation of the interrupt. As a result, the mailbox does not need to be read. AMCC Confidential and Proprietary DS1596 112 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book ADD-ON LOCAL BUS INTERFACE This chapter describes the Add-On Local bus interface of the S5920. The S5920 is designed to support connection to a variety of microprocessor buses and/or peripheral devices. The Add-On interface controls S5920 operation through the Add-On Operation Registers accessed through the 32 bit local bus. The Add-On local bus interface is synchronous to ADCLK. ADCLK is a 0-40 MHz clock input which can be configured as asynchronous to the PCI clock or synchronous when connected to the S5920 BPCLK output. The following sections describe the various interfaces to the PCI bus and how they are accessed from the Add-On bus. The DQMODE signal configures the data path width for all Add-On Operation register accesses, except for the Pass-Thru Data and Address registers. When DQMODE is low, DQ is configured as a 32-bit data bus. When DQMODE is high, DQ is configured as a 16-bit data bus. For 16-bit operation, BE3# is redefined as ADR1, providing an extra address input, and BE2# is unused. ADR1 selects the low or high words of the 32-bit S5920 Add-On Operation Registers. ADD-ON S5920 REGISTER ACCESSES The S5920 Add-On bus is very similar to that of a memory or peripheral device found in a microprocessor-based system. A 32-bit data bus with individual read and write strobes, a chip select and byte enables are provided. Register Access Signals ADD-ON INTERFACE SIGNALS The Add-On bus provides a number of system signals to allow Add-On logic to monitor PCI bus activity, to indicate status conditions (interrupts), and to configure the S5920 Add-On bus. SYSTEM SIGNALS BPCLK is a buffered version of the PCI clock. The PCI clock can operate from 0 MHz to 33 MHz. SYSRST# is a buffered version of the PCI reset signal, and may also be toggled by host application software through bit 24 of the Reset Control Register (RCR). IRQ# is the PCI interrupt request output to the Add-On bus. This signal is active low and can indicate multiple conditions. Add-On interrupts can be generated from the mailbox interface or to indicate start of BIST. The conditions which will generate an IRQ# due to mailbox activity are discussed in the mailbox chapter. The IRQ# output is deasserted when acknowledged by writing a 1 to the corresponding interrupt bit in the Add-On Interrupt Control/Status Register (AINT). See Table 3. The PTMODE signal (Pass-Thru Mode) controls the Pass-Thru interface only. Asserting it will configure the Pass-Thru in Passive mode and low will configure the Pass-Thru in Active mode. ADDINT# is an Add-On interrupt input pin. When asserted, it will cause the PCI interrupt output pin (INTA#) to assert. The ADDINT# is a level-sensitive input. Any number of Add-On peripheral interrupt sources can drive this input. There must be a pull-up resistor on the board to pull it high when inactive. This interrupt has to be enabled by Bit 13 of the INTCSR. It is the responsibility of the PCI host to clear the interrupt source of ADDINT# in order to have the pending interrupt deasserted. Register accesses to the S5920 Add-On Operation Registers are synchronous to the Add-On input clock (ADCLK). The following signals are required to complete a register access to the S5920. BE[3:0]# Byte Enable Inputs. These signals identify which bytes of the DQ bus are valid during AddOn bus transactions. BE0# indicates valid DQ[7:0], BE1# a valid DQ[15:8], etc. When DQ is configured for 16-bit operation, BE2# is not defined and BE3# becomes ADR1. ADR[6:2] A ddress Register Inputs. These pins address a specific Add-On Operation Register within the S5920. When DQ is configured for 16bit operation, an additional input, ADR1 is available to allow the 32-bit operation registers to be accessed in two 16-bit cycles. RD# Read Enable Input. WR# Write Enable Input. SELECT# Chip Select Input. This input indicates RD#, WR#, ADR[6:2] and BE[3:0] are valid. DQ[31:0] Bi-directional Data Bus. These I/O pins are the Add-On data bus. S5920 General Register Accesses For many Add-On applications, Add-On logic does not operate at the PCI bus frequency. This is especially true for Add-On designs implementing a microprocessor, which may be operating at a lower or higher frequency. The RD# and WR# inputs become enables, using ADCLK to clock data into and out of registers. All inputs are sampled on the rising edge of ADCLK. AMCC Confidential and Proprietary DS1596 113 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Figures 1 and 2 show basic operation register access timing relationships. Detailed AC timings are in Electrical and AC Characteristics. Chapter 10. For reads (Figure 1), data is driven onto the DQ bus on the ADCLK cycle after RD# is sampled asserted. When RD# is not asserted, the DQ outputs float. The address, byte enable, and RD# inputs must meet setup and hold times relative to the rising edge of ADCLK. For writes (Figure 2), data is clocked into an operation register on the rising edge of ADCLK in which WR# is sampled asserted. Address, byte enables, WR# and data must all meet setup and hold times relative to the rising edge or ADCLK. Figure 59. Read Operation Register ADCLK SELECT# ADR[6:2] BE[3:0]# RD# DQ[31:0] VALID VALID VALID S5920 16-bit Mode Register Accesses In the S5920 there are two methods of defining the Add-On DQ width: one is by using the DQMODE pin and the other is to define a Pass-Thru region size of 8, 16 or 32 bits. The DQMODE pin allows external 16-bit devices to access S5920 Operation Registers without additional logic. The external device is able to write and read the S5920 32-bit registers in two 16-bit cycle accesses. When performing an Operation Register a c c e s s w i t h t h e D Q M O D E p i n s e t f o r 1 6 b i ts (DQMODE = 1), only the lower half (DQ[15:0]) of the DQ bus is driven during a read or write. The S5920 internally steers the data bus and the byte enables based on the BE3# input. It is important to note that the DQMODE pin has no effect on accesses to the Pass-Thru Data Register. For non 32-bit Pass-Thru regions, the region size should be used instead. In 16bit mode, a 32-bit DWORD write is performed in two cycles: Cycle 1: D Q[15:0] is driven with the lower-WORD. WR#, ADR and SELECT# are asserted, BE[1:0]# indicates which bytes of the WORD are valid, and BE3# (which has been redefined as ADR1 when in 16-bit mode) is set to zero, indicating that the write is for the lower-WORD of the DWORD transfer. DQ[15:0] will be written to the bottom 16 bits of the internal 32-bit register (be it a Mailbox, Pass-Thru configuration register, etc.). Cycle 2: D Q[15:0] is driven with the upper-WORD. WR#, ADR and SELECT# are asserted, BE[1:0]# indicate which bytes of the WORD are valid, and BE3# is set to one, indicating that the write is for the upperWORD of the DWORD transfer. DQ[15:0] will be written to the upper 16-bits of the internal 32-bit register. Figure 60. Write Operation Register ADCLK SELECT# ADR[6:2] BE[3:0]# WR# Transfered DQ[31:0] DATA VALID VALID AMCC Confidential and Proprietary DS1596 114 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Figure 61. 16 Bit Mode Operation Register DWORD Write/Read 0 ADCLK DQMODE SELECT# ADR[6:2] BE3# BE[1:0]# WR# RD# DQ[15:0] 5678h 1234h 1234 5678h 00b 10b 01b 00b 60h 60h 1234 56789 10 11 12 13 Figure 3. 16 Bit Mode Operation Register DWORD Write/Read Figure 3 shows an example of a DWORD write of 12345678h using a 16 bit-mode write transfer, and is described as follows: Clock 1: ADR[6:2], BE[3:0], SELECT# and WR# are driven, DQ[15:0] is driven with the data to be written. BE3# is low indicating that the DQ bus data is to be written to the lower WORD of the register. BE[2:0]# is 00h, indicating that both bytes on the DQ[15:0] bus are valid and should be written to the register indicated by ADR[6:2]. Clock 2: The rising edge of clock 2 writes 5678h into the lower WORD of the register. 1234h is driven onto the DQ[15:0] bus. BE3# is driven high, indicating the DQ bus data is to be written to the upper WORD of the register. BE[2:0]# is 10h indicating that the lower byte of the WORD on DQ[15:0] bus is valid. This example shows how the BEs function. Clock 3: The rising edge of clock 3 writes 34h into the lower byte of the upper WORD of the register. BE[2:0]# is “01” indicating the upper byte on DQ[15:0] is valid. Clock 4: The rising edge of clock 4 writes 12h into the upper byte of the upper WORD of the register. 12345678h is in the register selected by ADR[6:2]. SELECT#, ADR[6:2], WR#, BE[3:0]# and DQ are deasserted. No read or write occurs on the rising edge of clocks 5 and 6. Figure 3 also shows a DWORD read of 12345678h from the same register, using a 16-bit mode read transfer, and is described as follows: Clock 6: ADR[6:2], SELECT# and RD# are asserted. BE3# is high, indicating the upper WORD of the register is to be driven onto DQ[15:0] and BE[1:0]# is 00h indicating both bytes of the WORD are to be driven. Clock 7: The S5920 drives 1234h onto DQ[15:0] as a result of the read issued during the previous cycle. BE3# is next driven low to indicate the lower WORD of the register is to be driven onto DQ[15:0]. BE[1:0]# is 00h indicating both bytes of the WORD should be driven onto DQ[15:0]. Note: in the event that BE[0]# was 1b, DQ[7:0] would NOT be driven during Clock 8, it would remain tri-state. The only exception to this is if ADR[6:2] indicated the Pass-Thru Data Register, where all of DQ[15:0] would be driven, regardless of the state of BE[1:0]#. Clock 8: On the rising-edge, Add-On logic latches data 1234h. The S5920 drives 5678h onto DQ[15:0] as a result of the read issued during the previous cycle. ADR[6:2], SELECT#, RD# and BE[3:0]# are deas serted, completing the transfer. Clock 9: On the rising-edge, Add-On logic latches data 5678h. DQ[15:0] returns to tri-state as RD# was sampled deasserted. AMCC Confidential and Proprietary DS1596 115 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book MAILBOX OVERVIEW For a detailed description of the Mailbox interface, reference Chapter 8. READ FIFO OVERVIEW The S5920 has an 8x32-bit Read FIFO, which allows data to be prefetched from the add on bus. The user can program the device to prefetch 2,4 or 8 DWORDs for each region or disable prefetching completely. For the first PCI read cycle, the device will request data from the Add-On bus. As the PCI bus reads the FIFO, and until after the PCI transfer has finished, the S5920 will prefetch the next N (2, 4 or 8) DWORDs from the Add-On. The prefetched data is valid as long as the PCI read addresses are sequential. If the current PCI read address is not the previous address plus four, or if a PCI write access occurs, the S5920 will flush the FIFO and start a new transfer at this address. Flushing the FIFO will incur a minimum loss of one PCI clock cycle or possibly more if the Add-On logic has not finished its current prefetching transfer. Note that prefetching is not performed past the upper limit of the base address region. In fact, prefetching is disabled when the PCI address is eight DWORDs from the end of the region. Prefetch cycles are always 32 bits regardless of AddOn bus width or the byte enables requested by the PCI. PASS-THRU OVERVIEW The S5920 provides data transfers between the PCI bus and the user local bus through the Pass-Thru data channel. Using a handshaking protocol with Add-On device(s), the PCI bus can directly access data on the Add-On bus and internal S5920 Operation registers. The Pass-Thru data channel is very flexible for user memory access or accessing registers within peripherals on the Add-On bus. Pass-Thru operation in Active or Passive mode requires an external non-volatile memory device to define and configure the Pass-Thru channel region sizes and bus widths. Four user-configurable Pass-Thru regions are available in the S5920. Each region is defined by a PCI Configuration Base Address Register (BADR1-4). A Pass-Thru region defines a block of predefined user space address in either host memory or I/O areas. Memory mapped regions can be requested below 1 Mbyte (Real Mode address space for a PC). Each region is configurable for bus widths of 8, 16 or 32 bits for the Add-On bus interface. The S5920 Pass-Thru channel supports single data transfers as well as burst transfers. When accessed with burst transfers, the S5920 supports data transfers at the full PCI bandwidth. The data transfer rate is only limited by the PCI initiator performing the access and the speed of the Add-On bus logic. FUNCTIONAL DESCRIPTION The S5920 Pass-Thru interface supports both single cycle (one data phase) and burst accesses (multiple data phases). Pass-Thru Transfers WRITE FIFO OVERVIEW For PCI write cycles, the S5920 has an 8x32-bit Write FIFO to increase performance for slow Add-On devices. When the FIFO is enabled, the S5920 will accept data transfers from the PCI bus at zero wait states until the FIFO is full. The device continues to fill the FIFO as long as the transfers are sequential. The S5920 can continue accepting sequential write PCI transfers as long as the FIFO is not full and the boundary of the Pass-Thru region defined by the Base Address Register is not crossed. If the next data access is for a non-sequential address, the FIFO must first be emptied by the Add-On peripheral in order for the next transfer to occur. The Write FIFO can be disabled, thus configuring the FIFO to act as a single DWORD data buffer. In this case, PCI Write Posting is not possible. The Pass-Thru interface offers two different modes of operation: Passive mode and Active mode. Passive mode is configured by strapping the pin PTMODE high, while Active mode is configured by strapping the pin PTMODE Low. PTMODE = 1 - Passive Operation PTMODE = 0 - Active Operation Passive operation allows external Add-On bus peripherals to provide read and write control signals to the S5920. The user drives SELECT#, RD#, WR#. ADR[6:2] and PTRDY#. The Add-On bus logic has the flexibility of determining when it wants to perform reads/writes. Some applications may require that a PCI address be passed for Pass-Thru accesses. For example, a 4Kbyte Pass-Thru region on the PCI bus may corre spond to a 4-Kbyte block of SRAM on the Add-On card. If a PCI initiator accesses this region, the AddOn would need to know the offset within the memory device to access. The Pass-Thru Address Register (APTA) allows Add-On logic to access address inforDS1596 116 AMCC Confidential and Proprietary S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book mation for the current PCI cycle. When the PCI bus performs burst accesses, the APTA register is incremented by the S5920 to reflect the address of the current data phase. PTNUM[1:0] is used to determine what region owns the current data access. For PCI writes to the Add-On, the S5920 transfers data from the PCI bus into the Pass-Thru Write FIFO. When the Pass-Thru write FIFO becomes not empty, the S5920 asserts the Pass-Thru status signals to indicate to the Add-On that data is present. The Add-On logic will then read data from the FIFO. The S5920 continues accepting write data from the PCI initiator as long as the 8x32 FIFO is not full. For PCI reads from the Add-On, the S5920 asserts the Pass-Thru status signals to indicate to the Add-On that data is required. The Add-On logic should write the requested data into the Pass-Thru Read FIFO. The S5920 will assert TRDY# to the PCI bus after the AddOn logic has transferred data into the FIFO. As long as data is in the FIFO, and PCI read data is still requested, TRDY# will continue to be asserted. If the Add-On cannot provide data quickly enough, the S5920 signals a disconnect to the PCI bus. This allows the PCI bus to perform other tasks, rather than waiting for a slow target. The S5920 will prefetch data if enabled. Signal Function PTATN# PTBURST# PTNUM[1:0] PTBE[3:0]# PTWR PTADR# This output indicates a Pass-Thru access needs servicing. This output indicates that the current Pass-Thru access is a PCI burst transfer or a single cycle transfer. PTBURST# is deasserted immediately after the second to last burst data has been transferred on the PCI side. PTBURST# is also active during prefetch cycles. These outputs indicate which Pass-Thru region decoded the PCI address. These outputs indicate which data bytes are valid (PCI writes), or requested (PCI reads). See timing diagrams for further details. PTBE0# = 0 Byte 0 is valid, PTBE1# = 0 Byte 1 is valid, PTBE# = 0 Byte 2 is valid, PTBE3# = 0 Byte 3 is valid. This output indicates if the Pass-Thru access is a PCI read or a write. When asserted, this pin drives the Pass-Thru Address Register contents onto the Add-On data bus. This input enables the DQ[31:0] data bus to become active immediately. There is NO pipeline delay from PTADR# to DQ, as there is from RD# to DQ. As result, this is an asychronous input. In Passive mode, this input indicates the current Pass-Thru transfer has been has been completed by the Add-On. In Active mode, this input indicates that wait states are to be inserted for the next transfer. Input Add-On clock (to synchronize Pass-Thru data register accesses). PTRDY# ADCLK Pass-Thru Status/Control Signals The S5920 Pass-Thru registers are accessed using the Add-On register access pins. The Pass-Thru Address Register (APTA) can, optionally, be accessed using a single, direct access input, PTADR#. PassThru cycle status indicators are provided to control Add-On logic based on the type of Pass-Thru access occurring (single cycle, burst, etc.). The signals in the table above are provided for Pass-Thru operation: ated by the PCI bus. The Pass-Thru interface is designed to allow Add-On logic to function without knowledge of PCI bus activity. Add-On logic only needs to react to the Pass-Thru status signals. The S5920 PCI device independently interacts with the PCI initiator to control data flow between the devices. The following sections describe the PCI and Add-On bus interfaces. The PCI interface description provides a basic overview of how the S5920 interacts with the PCI bus, and may be useful in system debugging. The Add-On interface description indicates functions required by Add-On logic and details the Pass-Thru handshaking protocol. PCI Bus Interface BUS INTERFACE The Pass-Thru data channel allows PCI initiators to read or write to resources on the Add-On bus. A PCI initiator may access the Add-On with single data phase cycles or multiple data phase bursts. The AddOn interface implements Pass-Thru status and control signals used by logic to complete data transfers initiAMCC Confidential and Proprietary The S5920 device examines all PCI bus cycle addresses. If the address associated with the current DS1596 117 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book cycle decodes to one of the S5920 Pass-Thru regions, DEVSEL# is asserted. If the Pass-Thru logic is currently idle (not busy finishing a previous Pass-Thru operation), the bus cycle type is decoded and the AddOn Pass-Thru status outputs are set to initiate a transfer on the Add-On bus. The following sections describe the behavior of the PCI interface for PassThru accesses to the S5920. Single cycle accesses, burst accesses, and target-initiated retries are detailed. PCI Pass-Thru Single Cycle Accesses burst access, the Pass-Thru status indicators notify the Add-On logic. For Pass-Thru burst writes, the S5920 responds immediately (asserting TRDY#). The S5920 transfers the first data phase of the burst into the FIFO, and stores the PCI address in the Pass-Thru Address Register (APTA). The S5920 can accept up to 8 DWORDs from the PCI bus before transferring one DWORD on the Add-On side. If the Add-On bus is slow, the device will keep the FIFO full until the data is ready to be transferred by the slow Add-On bus. If the Add-On bus is fast at accepting the data, then the FIFO will continue an indefinite burst, or until the PCI master is forced to relinquish the bus for arbitration reasons, or the PCI bus master has gone beyond the Pass-Thru region address space. For burst accesses, the APTA is automatically incremented by the S5920 for each data phase. For Pass-Thru burst reads, the S5920 claims the PCI cycle (asserting DEVSEL#). The request for data is passed on to Add-On logic and the PCI address is stored in the APTA register. The device will prefetch data if the feature is enabled. The S5920 then drives the requested data on the PCI bus and asserts TRDY# to begin the next data phase. The APTA register is automatically incremented by the S5920 after each data phase. PCI Disconnect Conditions A single cycle transfer is the simplest of PCI bus transactions. Single cycle transfers have an address phase and a single data phase. The PCI bus transaction starts when an initiator drives address and command information onto the PCI bus and asserts FRAME#. The initiator always deasserts FRAME# before the last data phase. For single cycle transfers, FRAME# is only asserted during the address phase (indicating the first data phase is also the last). When the S5920 sees FRAME# asserted, it samples the address and command information to determine if the bus transaction is intended for it. If the address is within one of the defined Pass-Thru regions or internal PCI Operation Register, the S5920 accepts the transfer (asserts DEVSEL#), and stores the PCI address in the Pass-Thru Address Register (APTA). For Pass-Thru writes, the S5920 responds immedi ately (asserting TRDY#) and transfers the data from the PCI bus into the write FIFO as long as the write FIFO is not full. The S5920 then indicates to the AddOn interface that a Pass-Thru write is taking place and waits for Add-On logic to complete the transfer. Once the S5920 has captured the data from the PCI bus, the transfer is finished from the PCI bus perspective, and the PCI bus becomes available for other transfers. For Pass-Thru reads, the S5920 indicates to the AddOn interface that a Pass-Thru read is taking place and waits for Add-On logic to complete the cycle. If the Add-On cannot complete the cycle quickly enough, the S5920 requests a retry from the initiator. The S5920 will fetch one DWORD from the Add-On side, and store it in the Read FIFO. PCI Pass-Thru Burst Accesses Before discussing what causes the S5920 to issue a disconnect on the PCI bus, it might be useful to distinguish between a disconnect and retry. A retry occurs when a PCI initiator does not receive a single TRDY#, but is issued a STOP# instead. In this case, no data is transferred. The PCI 2.1 spec states that the initiator is required to come back and complete this transfer. A disconnect occurs after at least one data phase was completed (TRDY# and IRDY# asserted simulta neously). This occurs when a STOP# is asserted either with a TRDY# or after a TRDY#/IRDY# transfer. In this case, the initiator is not required to return to complete the transfer. In some applications, Add-On logic may not be able to respond to Pass-Thru accesses quickly. In this situation, the S5920 will Retry the cycle on the PCI side. For PCI write cycles, the S5920 will accept up to 8 DWORDs without a disconnect or until the FIFO is full. For a PCI read cycle, the first access needs to take less than 16 PCI clocks, otherwise the device will issue a Retry. A subsequent read transfer must take less than 8 PCI clocks, otherwise the device will issue a disconnect. For PCI Pass-Thru burst accesses, the S5920 captures the PCI address and determines if it falls into one of the defined Pass-Thru regions. Accesses that fall into a Pass-Thru region or internal PCI Operation Register are accepted by asserting DEVSEL#. The S5920 monitors FRAME# and IRDY# on the PCI bus to identify burst accesses. If the PCI initiator is performing a AMCC Confidential and Proprietary DS1596 118 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book With many devices, particularly memories, the first access takes longer than subsequent accesses (assuming they are sequential and not random). For this reason, the PCI specification allows 16 clocks to respond to the first data phase of a PCI cycle and 8 clocks for subsequent data phases (in the case of a burst) before a retry/disconnect is issued by the S5920. The S5920 also requests a disconnect if an initiator attempts to burst past the end of a Pass-Thru region. The S5920 updates the Pass-Thru Address Register (APTA) for each data phase during bursts, and if the updated address is not within the current Pass-Thru region, a disconnect is issued. Accesses to undefined addresses will cause the PCI host to receive a Master Abort cycle (no DEVSEL# is asserted by the S5920). For example, a PCI system may map a 512 byte PassThru memory region to 0DC000h to 0DC1FFh. A PCI initiator attempts a four DWORD burst with a starting address of 0DC1F8h. The first and second data phases complete (filling the DWORDs at 0DC1F8h and 0DC1FCh), but the third data phase causes the S5920 to issue a disconnect. This forces the initiator to present the address 0DC200h on the PCI bus. If this address is part of another S5920 Pass-Thru region, the device accepts the access, but if not, a Master Abort cycle occurs. PCI Write Disconnect Add-On cannot complete the access within 16 clocks, a retry is requested (STOP# asserted without data transfer). If the Add-On manages to complete the data transfer into the PT Read FIFO, but after a retry was issued, the data is held in the FIFO until the original master comes back to read it. All subsequent PCI accesses to a Pass-Thru address other than the one corresponding to the data in the FIFO will be terminated with a PCI retry. Only a PCI access with a matching address can access the data in the PT Read FIFO, and thus release the Pass-Thru region for other accesses. If the Add-On is busy performing a Pass-Thru write operation when a PCI read occurs, the S5920 requests an immediate retry. If the Add-On is busy performing a Pass-Thru read operation when another PCI read occurs, the S5920 determines whether the read is a retry from a previous access, and if so, attempts to continue the read where it left off. If the address is non-sequential, the new access is issued a retry. This allows the PCI bus to perform other operations. S5920 PCI Operation Registers may be accessed while the Add-On is still completing a Pass-Thru access. Only other Pass-Thru region accesses receive retry requests. If the prefetch feature is enabled, the Pass-Thru interface will prefetch data, which should improve the performance on subsequent cycles to the same region. In the event that the Add-On cannot prefetch the first data before the S5920 issues a PCI retry, the prefetched data will be held in the read FIFO until the original master comes back to request it. Other PCI read requests to the Pass-Thru region will be terminated with immediate Retries. If the prefetch feature is disabled, a PCI read cycle is not completed until the data is first transferred from the Add-On bus into the PT Read FIFO. The device will not prefetch, but will only request data from the AddOn bus after the PCI bus has requested the data. Pass-Thru bursts will not be performed in this case. In the event that a non-prefetchable Add-On cannot provide the second (or third, or fourth...) data to the PCI read request within the PCI Target Subsequent Latency period (eight PCI clocks), the S5920 will issue a PCI disconnect (STOP# asserted with data transfer). If the Add-On manages to transfer the second (or third, or fourth...) data to the PT Read FIFO, but after the disconnect, the data may be held in the FIFO until the original master comes back to read it. Depending upon the setting of the Retry Flush Enb bit, the data is held in the FIFO, and all other PCI read requests will be terminated with immediate Retries. When the S5920 issues a disconnect for a PCI PassThru write, it indicates that the Add-On is still completing a previous non-sequential Pass-Thru access or the FIFO is full. If the incoming access is a continuation of a previous one, no disconnect is issued and the transaction can continue where it left off (perhaps due to a previous disconnect or master time-out). PCI Operation Registers may be accessed while the Add-On is still completing a Pass-Thru access. Only Pass-Thru region accesses receive disconnect requests. PCI Read Disconnect If the S5920 issues a disconnect for a PCI Pass-Thru read, this indicates that the Add-On could not complete the read in the required time (16 clocks for the first data phase, 8 PCI clocks for the 2nd or later data phases). When the PCI performs a read to a Pass-Thru region, the Add-On device must complete a Pass-Thru data transfer by writing the appropriate data into the PassThru Data FIFO (APTD). If the Add-On can perform this before the required time (see above), the S5920 asserts TRDY# to complete a PCI read transfer. If the AMCC Confidential and Proprietary DS1596 119 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book S5920 PASSIVE MODE OPERATION The Pass-Thru address and data registers can be accessed as Add-On operation registers. The PassThru FIFO is updated on the rising edge of ADCLK. For this reason, all Pass-Thru inputs must be synchronous to ADCLK. In the following sections the Add-On Pass-Thru interface is described for Pass-Thru single cycle accesses, burst accesses, target-requested retries, and when using 8-bit and 16-bit Add-On data buses. Single-Cycle PCI to Pass-Thru Write PTBE[3:0]# 0h. Indicates the Pass-Thru access has all bytes valid. Clock 2: SELECT#, ADR[6:2] and BE[3:0]# inputs are driven to read the Pass-Thru Write FIFO at offset 2Ch. DQ[31:0] is driven one clock after RD# and SELECT# are asserted. PTRDY# is asserted, indicating that the transfer is complete. Clock 3: PTBE[3:0] will update one clock after RD# is asserted to indicate which bytes have not yet been read. The data is also driven on the DQ bus since RD# was asserted a clock earlier. Since PTRDY# was sampled asserted, PTATN# is deasserted and the PassThru access is complete. If the Add-On logic requires more time to complete the read, PTRDY# can be delayed, extending the Pass-Thru cycle. Clock 4: A s PTATN# is deasserted, the Pass-Thru access is complete, and the S5920 can accept new Pass-Thru accesses starting on the next clock. The other Pass-Thru signals can also change state (in anticipation of a new transfer). The S5920 stops driving the DQ bus as RD# and SELECT# were not valid on the previous cycle. A single-cycle Pass-Thru write operation occurs when a PCI initiator writes a single value to the Pass-Thru region. PCI single cycle transfers consist of an address phase followed by one data phase. During the address phase of the PCI transfer, the S5920 stores the PCI address into the Pass-Thru Address Register (APTA). If the S5920 determines that the address is within one of its defined Pass-Thru regions, it captures the PCI data into the FIFO. Figure 4 shows a single cycle Pass-Thru write access in the Passive Mode. The Add-On must read the data stored in the FIFO and transfer it to its destination. If the proper SELECT#, ADR[6:2] and BE[3:0]# signals are present, the S5920 will drive data one clock after RD# is asserted. It will stop driving data after the rising edge of ADCLK when RD# has been sampled deasserted. Clock 0: T he PCI bus cycle address information is stored in the S5920 and later stored in the Pass-Thru Address Register. The PCI address is recognized as a write to Pass-Thru region 1. The PCI data is stored in the S5920 Write FIFO. Clock 1: P ass-Thru signals PTATN#, PTBURST#, PTNUM[1:0], PTWR and PTBE[3:0] are driven to indicate what action is required by Add-On logic. These status signals are valid only when PTATN# is active. Add-On logic can decode status signals upon the assertion of PTATN#. Figure 5 shows a single cycle Pass-Thru write for the Passive Mode using the Pass-Thru address information. This provides PCI cycle address information to select a specific address location within an Add-On memory or peripheral. Add-On logic may latch the address for use during the data transfer (if PTADR# was asserted). Typically, the entire 32-bit address is not required. The Add-On may implement a scheme where only the required number of address bits are latched. It may also be useful to use the Pass-Thru region identifiers, PTNUM[1:0], as address lines. For example, Pass-Thru region 1 might be a 64K block of SRAM for data, while Pass-Thru region 2 might be 64K of SRAM for code storage (downloaded from the host during initialization). Using PTNUM0 as address line A16 allows two unique add-on memory regions to be defined. Unlike all other Add-On operation register reads, the Add-On PTADR# input directly accesses the PassThru Address Register and drives the contents onto the data bus during the same clock cycle. WR# must not be asserted the same time as PTADR#, or there would be contention on the DQ bus! However, it is permitted to assert RD# and PTADR# during the same cycle. This is because all reads performed with RD# are pipelined, while address reads with PTADR# are not pipelined. PTATN# Asserted. Indicates a Pass-Thru access is pending PTBURST# Deasserted. The access has a single data phase. PTNUM[1:0] 1h. Indicates the access is to Pass-Thru region 1. PTWR Asserted. The Pass-Thru access is a write. AMCC Confidential and Proprietary DS1596 120 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Figure 62. PCI To Add-On Passive Write 0 ADCLK PTATN# PTBURST# PTNUM[1:0] PTWR PTBE[3:0] SELECT# ADR[6:2] BE[3:0]# RD# DQ[31:0] PTRDY# PTDATA 2Ch 0h 0h Fh 1h 1 2 3 4 5 PTWR and PTBE[3:0] will update on the next rising edge of ADCLK. Clock 1: P ass-Thru signals PTATN#, PTBURST#, PTNUM[1:0], PTWR and PTBE[3:0] are driven to indicate what action is required by Add-On logic. These status signals are valid only when PTATN# is active. Add-On logic can decode status signals upon the assertion of PTATN#. PTATN# Asserted. Indicates Pass-Thru access is pending. PTBURST# Not asserted. The access has a single data phase. PTNUM[1:0] 1h. Indicates the access is to Pass-Thru region 1. PTWR Asserted. The Pass-Thru access is a write. PTBE[3:0]# 0h. Indicate the Pass-Thru has all bytes valid. Clock 2: The PTADR# input is asserted to read the Pass-Thru Address Register. The assertion of PTADR# will immediately cause the address to be driven on the DQ bus. RD#, SELECT#, byte enable, and the address inputs are asserted to read the PassThru Data Register at offset 2Ch. DQ[31:0] is driven one clock after RD# and SELECT# are asserted. Asserting PTADR# and RD# at the same time will save a clock cycle, since the assertion of the RD# won’t cause the data to be driven until a clock later. The Add-On also asserts PTRDY#, indicating that the current transfer is complete. Clock 3: PTBE[3:0] are updated to indicate which bytes have not yet been read. Data is driven on the DQ bus because RD# was asserted a clock earlier. The Add-On logic reads the data and deasserts PTRDY#. As PTRDY# was sampled asserted, PTATN# is immediately deasserted and the Pass-Thru access is completed with the next clock. If add-on logic requires more time to read the Pass-Thru Data Register (slower memory or peripherals), PTRDY# can be delayed, extending the cycle. Clock 4: A s PTATN# is deasserted, the Pass-Thru access is complete, and the S5920 can accept new Pass-Thru accesses starting on the next clock. The other Pass-Thru signals can also change (in anticipation of a new transfer). The S5920 stops driving the DQ bus as RD# and SELECT# were not valid on the previous cycle. Single-Cycle PCI to Pass-Thru Read Figure 63. PCI To Add-On Passive Write w/Pass-Thru Address 0 ADCLK PTATN# PTBURST# PTNUM[1:0] PTWR PTBE[3:0] SELECT# ADR[6:2] BE[3:0]# RD# DQ[31:0] PTADR# PTRDY# ADDR DATA 2Ch 0h 0h Fh 1h 1 2 3 4 Clock 0: The address is recognized as a PCI write to Pass-Thru region 1. The PCI bus write address is stored in the Pass-Thru Address Register. The PCI bus write data is stored in the S5920 Write FIFO. AddOn bus signals PTATN#, PTBURST#, PTNUM[1:0], AMCC Confidential and Proprietary A single-cycle PCI to Pass-Thru read operation occurs when a PCI initiator reads a single value from a PassThru region. PCI single cycle transfers consists of an DS1596 121 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book address phase followed by a single data phase. If the S5920 determines that the address is within one of its defined Pass-Thru regions, it indicates to the Add-On a write to the Pass-Thru Data Register (APTD) is required. Figure 6 shows a Passive Mode single cycle PassThru read access (Add-On write) using PTADR#. The Add-On reads data from a source on the Add-On and writes it to the APTD register. Figure 64. PCI To Add-On Passive Read 0 ADCLK PTATN# PTBURST# PTNUM[1:0] PTWR PTBE[3:0] SELECT# ADR[6:2] BE[3:0]# WR# DQ[31:0] PTRDY# PTADR# ADDR DATA 2Ch 0h 0h Fh 2h PTATN# Asserted. Indicates a Pass-Thru access is pending. PTBURST# Deasserted. The access has a single data phase. PTNUM[1:0] 2h. Indicates the access is to Pass-Thru region 2. PTWR Deasserted. The Pass-Thru access is a read. PTBE[3:0]# 0h. Indicates the Pass-Thru access has all bytes valid. Clock 2: The PTADR# input is asserted to read the Pass-Thru Address Register. The address can be latched on the next rising-edge of ADCLK. Clock 3: This turn-around cycle is required to avoid contention on the DQ bus. Time must be allowed after PTADR# is deasserted for the DQ outputs to float before add-on logic attempts to write to the Pass-Thru Read FIFO. Clock 4: WR#, SELECT#, BE[3:0]#, and ADR[6:2] are asserted to write to the Pass-Thru Read FIFO at address 2Ch. The Add-On logic drives the DQ bus with the requested data. PTRDY# is also asserted, indicating that the Add-On is finished with the transfer. Clock 5: The data on the DQ bus is latched into the Pass-Thru Read FIFO. As the S5920 samples PTRDY# asserted, PTATN# is deasserted and the Pass-Thru access is complete. PTBE[3:0] will update one clock after WR# is asserted to indicate which bytes have not yet been read. If add-on logic requires more time to provide data (slower memory or peripherals), PTRDY# can be delayed, extending the cycle. Clock 6: A s PTATN# is deasserted, the Pass-Thru access is complete, and the S5920 can accept new Pass-Thru accesses starting on the next clock. The other Pass-Thru signals can also change (in anticipation of a new transfer). PCI to Pass-Thru Burst Writes 1 2 3 4 5 6 Clock 0: The address is recognized as a PCI read of Pass-Thru region 2. The PCI bus read address is stored in the Pass-Thru Address Register. Add-On bus signals PTATN#, PTBURST#, PTNUM[1:0], PTWR and PTBE[3:0] will update on the next rising edge of ADCLK. Clock 1: P ass-Thru signals PTATN#, PTBURST#, PTNUM[1:0], PTWR and PTBE[3:0] are driven to indicate what action is required by Add-On logic. These status signals are valid only when PTATN# is active. Add-On logic can decode status signals upon the assertion of PTATN#. A PCI to Pass-Thru burst write operation occurs when a PCI initiator writes multiple values to a Pass-Thru region. A PCI burst cycle consists of an address phase followed by multiple data phases. If the S5920 determines that the requested address is within one of its defined Pass-Thru regions, the initial PCI address is stored into the Pass-Thru Address Register (APTA). The data from each data-cycle is individually latched into the Pass-Thru Data register (APTD) or Pass-Thru Write FIFO. AMCC Confidential and Proprietary DS1596 122 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Figure 65. PCI to Add-On Passive Burst Write 0 ADCLK PTATN# PTBURST# PTNUM[1:0] PTWR PTBE[3:0] SELECT# ADR[6:2] BE[3:0]# RD# DQ[31:0] PTADR# PTRDY# ADDR DATA1 DATA2 DATA3 DATA4 0h 0h 2Ch 0h 0h D1 D2 D3 D4 Fh 3h 1 2 3 4 5 6 7 8 9 10 11 12 13 Figure 7 shows a Passive mode PCI to Add-On burst write of four DWORDs. In the following example, AddOn logic incorporates the use of PTADR# followed by multiple data reads to the S5920. If Add-On logic does not support burst accesses, PTADR# can be pulsed for individual data reads. The S5920 automatically increments the address in the APTA register during PCI bursts. In this example PTRDY# is continually asserted, indicating that Add-On logic is capable of accepting one DWORD per clock cycle. In addition, the PTBE[3:0] signals indicate a unique byte-enable for each data transfer. The Pass-Thru Write FIFO (or APTD) can be disabled for bursts (do not accept PCI posted writes). In this case, the PCI is allowed to write to only one FIFO location and cannot continue bursting until the add-on has read the data. PTBURST# is never asserted when the PCI write FIFO is disabled. For this example, the Write FIFO is enabled. Clock 0: The address is recognized as a PCI write to Pass-Thru region 1. The PCI bus write address is stored in the Pass-Thru Address Register. The PCI bus write data is stored in the S5920 write FIFO. AddOn bus signals PTATN#, PTBURST#, PTNUM[1:0], PTWR and PTBE[3:0] will update on the next rising edge of ADCLK. Clock 1: P ass-Thru signals PTATN#, PTBURST#, PTNUM[1:0], PTWR and PTBE[3:0] are driven to indicate what action is required by Add-On logic. These status signals are valid only when PTATN# is active. Add-On logic can decode status signals upon the assertion of PTATN#. PTATN# Asserted. Indicates Pass-Thru access is pending. PTBURST# Asserted. The access has multiple data phases. PTNUM[1:0] 3h. Indicates the access is to Pass-Thru region 3. PTWR Asserted. Indicates the access is a write. PTBE[3:0]# D1. Indicates valid bytes for the first data transfer. Clock 2: The PTADR# input is asserted to read the Pass-Thru Address Register. The RD#, BE#, ADR[6:2] and SELECT# inputs are driven during this clock to read the Pass-Thru Data Register contents onto the DQ bus during the next clock. PTRDY# is asserted, indicating that the first transfer is complete. AMCC Confidential and Proprietary DS1596 123 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Clock 3: The Add-On latches the address. Data 1 is driven on the DQ bus as a result of the previous read. As PTRDY# is sampled asserted, the PTBE# outputs are updated to indicate which bytes are valid for the second transfer. The BE[3:0]#, ADR[6:2], and SELECT# inputs remain driven along with RD# to read out the next data. PTRDY# remains asserted, indicating that the second transfer is complete. Clock 4: A dd-On logic uses the rising edge of this clock to store DATA1. DATA2 is driven on the DQ bus as a result of the previous read. As PTRDY# is sampled asserted, the PTBE# outputs are updated to indicate which bytes are valid for the third transfer. PTRDY# remains asserted, indicating that the third transfer is complete. Clock 5: A dd-On logic uses the rising edge of this clock to store DATA2. PTBURST# is deasserted indicating that only a single data phase remains. DATA3 is driven on the Add-On bus. The PTBE# outputs are updated to indicate which bytes are valid for the last transfer. PTRDY# remains asserted, indicating that the current transfer is complete. Clock 6: A dd-on logic uses the rising edge of this clock to store DATA3 from the S5920. PTRDY# sampled completes the last data phase. As a result, the S5920 deasserts PTATN#, and drives DATA4 onto the DQ bus. As the Add-on sampled PTBURST# deasserted and PTATN# asserted, it recognizes that the previous read was the last one. As a result, the Add- On deasserts SELECT#, ADR[6:2], BE[3:0]#, RD# and PTRDY#. Clock 7: The Add-on logic stores DATA4 on the rising edge of this clock. As PTATN# is deasserted, the Pass-Thru access is complete, and the S5920 can accept new Pass-Thru accesses starting on the next clock. The other Pass-Thru signals can also change state (in anticipation of a new transfer). F igure 8 illustrates a Passive mode transfer with a burst of five DWORDs in a PCI to Pass-Thru burst write with PTRDY# used to insert wait states. In some applications, Add-On logic may not be required to transfer data on every ADCLK and can use PTRDY# to control the data rate transfer. In this example, AddOn logic latches data every other clock cycle. RD# is shown deasserted when PTRDY# is deasserted, but could remain active during the entire Add-On burst. In this case, the DQ would not go to tri-state between reads, and the PTBE# outputs would lose some of their significance (as they would transition one cycle early as a result of the “unused” read). Clock 0: The address is recognized as a PCI write to Pass-Thru region 0. The PCI bus write address is stored in the Pass-Thru Address Register. The PCI bus write data is stored in the S5920 write FIFO. AddOn bus signals PTATN#, PTBURST#, PTNUM[1:0], PTWR and PTBE[3:0] will update on the next rising edge of ADCLK. AMCC Confidential and Proprietary DS1596 124 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Figure 66. PCI to Add-On Passive Burst Write Using PTRDY# to assert Wait-States 0 ADCLK PTATN# PTBURST# PTNUM[1:0] PTWR PTBE[3:0] SELECT# ADR[6:2] BE[3:0]# RD# DQ[31:0] PTADR# PTRDY# ADDR DATA1 DATA 2 DATA 3 DATA 4 DATA 5 2Ch 0h D1 D2 D3 D4 D5 Fh 0h 1 2 3 4 5 6 7 8 9 10 11 12 13 Figure 67. PCI to Add-On Passive Burst Read Access 0 ADCLK PTATN# PTBURST# PTNUM[1:0] PTWR PTBE[3:0] SELECT# ADR[6:2] BE[3:0]# WR# DQ[31:0] PTRDY# PTADR# PTADDR DATA1 DATA2 DATA3 DATA4 2Ch 0h D1 D2 D3 D4 Fh 1h 1 2 3 4 5 6 7 8 9 AMCC Confidential and Proprietary DS1596 125 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Clock 1: P ass-Thru signals PTATN#, PTBURST#, PTNUM[1:0], PTWR and PTBE[3:0] are driven to indicate what action is required by Add-On logic. These status signals are valid only when PTATN# is active. Add-On logic can decode status signals upon the assertion of PTATN#. the next data transfer. RD# is also asserted, requesting DATA3 to be driven during the next clock cycle. Clock 8: PTRDY# is sampled asserted, thus completing the current data-phase. DATA3 is driven on the DQ bus, a result of a read during the previous cycle. The PTBE# outputs are updated to indicate which bytes are valid for the fourth transfer. Add-On logic is not fast enough to store the next data, so a wait state is activated by deasserting PTRDY#. RD# is also deasserted. Clock 9: A dd-On logic uses the rising edge of this clock to store DATA3. PTRDY# is sampled deas serted, so a wait state is activated. PTRDY# is asserted to indicate that the Add-On is ready to accept the next data transfer. RD# is also asserted, requesting DATA4 to be driven during the next clock cycle. Clock 10: PTRDY# is sampled asserted, thus completing the current data-phase. PTBURST# is deasserted, indicating that only one DWORD is left for transfer. DATA4 is driven on the Add-On DQ bus, a result of a read during the previous clock cycle. The PTBE# outputs are updated to indicate which bytes are valid for the last transfer. Add-On logic is not fast enough to store the next data, so a wait state is activated by deasserting PTRDY#. RD# is also deasserted. Clock 11: Add-On logic uses the rising edge of this clock to store DATA4. PTRDY# is sampled deas serted, so a wait state is activated. PTRDY# is asserted to indicate that the add-on is ready to accept the last data transfer. The add-on knows this is the last transfer as it has sampled PTBURST# deasserted and PTATN# asserted. RD# is also asserted, requesting DATA5 to be driven during the next clock cycle. Clock 12: PTRDY# is sampled asserted, indicating that the last transfer was completed. As a result, PTATN# is deasserted. As the Add-On has also finished its transfer, it deasserts RD#, SELECT#, BE[3:0]#. The last data, DATA5, is driven on the AddOn DQ bus. Clock 13: Add-On logic uses the rising edge of this clock to store DATA5. As PTATN# is deasserted, the Pass-Thru access is complete, and the S5920 can accept new Pass-Thru accesses starting on the next clock. The other Pass-Thru signals can also change state (in anticipation of a new transfer). Pass-Thru Burst Reads PTATN# Asserted. Indicates Pass-Thru access is pending. PTBURST# Asserted. The access has multiple data phases. PTNUM[1:0] 0h. Indicates the access is to Pass-Thru region 3. PTWR Asserted. Indicates the access is a write. PTBE[3:0]# D1. Indicates valid bytes for the first data transfer. C l o ck 2 : A d d - O n l o g i c s a mp l e s PTATN # a nd PTBURST# asserted, indicating the start of a burst. The Add-On asserts PTADR# to read the Pass-Thru Address Register. As it is not ready to receive any data yet, it does not initiate a data read. Clock 3: A dd-on logic latches the address. RD#, BE[3:0]#, ADR[6:2], and SELECT# inputs are asserted to select the Pass-Thru Data Register during the next clock. PTRDY# is also asserted to indicate the completion of the first data phase. Clock 4: As the S5920 sampled PTRDY# asserted, the first data phase is completed DATA1 is driven on the DQ bus, a result of the read from the previous clock cycle. The PTBE# outputs are updated to indicate which bytes are valid for the second transfer. Add-on logic is not fast enough to store the next data, so a wait state is activated by deasserting PTRDY#. RD# is also deasserted. Clock 5: A dd-On logic uses the rising edge of this clock to store DATA1. PTRDY# is sampled deas serted, so a wait state is activated. PTRDY# is asserted to indicate that the Add-On is ready to accept the next data transfer. RD# is also asserted, requesting DATA2 to be driven during the next clock cycle. Clock 6: PTRDY# is sampled asserted, thus completing the current data-phase. DATA2 is driven on the DQ bus, a result of a read during the previous clock cycle. The PTBE# outputs are updated to indicate which bytes are valid for the third transfer. Add-on logic is not fast enough to store the next data, so a wait state is activated by deasserting PTRDY#. RD# is also deasserted. Clock 7: A dd-On logic uses the rising edge of this clock to store DATA2. PTRDY# is sampled deas serted, so a wait state is activated. PTRDY# is asserted to indicate that the Add-On is ready to accept AMCC Confidential and Proprietary A Pass-Thru burst read operation occurs when a PCI initiator reads multiple DWORDs from a Pass-Thru region. A burst transfer consists of a single address and multiple data phases. The S5920 stores the PCI DS1596 126 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book address into the Pass-Thru Address Register (APTA). If the S5920 determines that the address is within one of its defined Pass-Thru regions, it indicates to the Figure 68. PCI to Add-On Passive Burst Read 0 ADCLK PTATN# PTBURST# PTNUM[1:0] PTWR PTBE[3:0]# SELECT# ADR[6:2] BE[3:0]# WR# DQ[31:0] PTADR# PTRDY# ADDR Data 1 Data 2 Data 3 Data 4 2Ch 0h D1 D2 D3 D4 Fh 0h 1 2 3 4 5 6 7 8 9 10 11 12 13 Add-On that a write to the Pass-Thru Data Register or Pass-Thru Read FIFO (APTD) is required. Figure 9 shows a Passive Mode Pass-Thru burst read access (Add-On write) of four DWORDs, using PTADR# to provide an address-phase. Clock 0: P CI address information is stored in the Pass-Thru Address Register. The address is recognized as a PCI read of Pass-Thru region 1. Add-On bus signals PTATN#, PTBURST#, PTNUM[1:0], PTWR and PTBE[3:0] will update on the next rising edge of ADCLK Clock 1: P ass-Thru signals PTATN#, PTBURST#, PTNUM[1:0], PTWR and PTBE[3:0] are driven to indicate what action is required by Add-On logic. These status signals are valid only when PTATN# is active. Add-On logic can decode status signals upon the assertion of PTATN#. PTNUM[1:0] 1h. Indicates the access is to Pass-Thru region 1. PTWR Deasserted. Indicates the access is a read. PTBE[3:0]# D1. Indicates valid bytes for the first data transfer. Clock 2: The Add-On logic has sampled PTATN# and PTBURST# active, indicating that at least two read data transfers are requested by the PCI. The Add-On will start servicing the Burst Read transfer by first reading the Pass-Thru Address via the PTADR# input. This is an asynchronous read, meaning that the address will appear on DQ after a propagation delay from the assertion of PTADR#. In the event that the address is not required, this cycle and the next could be skipped (as the next clock provides a turn-around cycle). Clock 3: The Add-On logic will latch the Pass-Thru address on the rising edge of this clock. This cycle is also required to avoid contention on the DQ bus. Time must be allowed after PTADR# is deasserted for the PTATN# Asserted. Indicates Pass-Thru access is pending. PTBURST# Asserted. The access has multiple data phases. AMCC Confidential and Proprietary DS1596 127 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book DQ outputs to float before Add-On logic attempts to write to the Pass-Thru Read FIFO. Clock 4: T he BE[3:0]#, ADR[6:2], and SELECT# inputs are asserted. WR# and DQ are asserted, indicating that DATA1 is to be written to the PT Read FIFO on the next clock. PTRDY# is asserted, to indicate the completion of the current data phase. Clock 5: A s the S5920 samples WR# asserted, it writes DATA1 into the PT Read FIFO. PTRDY# is sampled asserted, which completes the first data transfer and updates the internal FIFO pointers. The PTBE# outputs are updated to indicate which bytes are valid for the second transfer. The Add-On logic samples PTBURST# asserted, so it knows more data is being requested. The Add-On keeps WR# asserted, and drives DATA2 onto the DQ bus. PTRDY# is also asserted to complete the current data phase. Clock 6: A s the S5920 samples WR# asserted, it writes DATA2 into the PT Read FIFO. PTRDY# is sampled asserted, which completes the second data transfer and updates the internal FIFO pointers. The PTBE# outputs are updated to indicate which bytes are valid for the third transfer. The Add-On logic samples PTBURST# asserted, so it knows more data is being requested. The Add-On keeps WR# asserted, and drives DATA3 onto the DQ bus. PTRDY# is also asserted to complete the current data phase. Clock 7: A s the S5920 samples WR# asserted, it writes DATA3 into the PT Read FIFO. PTRDY# is sampled asserted, which completes the third data transfer and updates the internal FIFO pointers. The PTBE# outputs are updated to indicate which bytes are valid for the last transfer. The S5920 deasserts PTBURST#, indicating that the previous read was the second to last. The next transfer from the Add-On bus will be the last. The Add-On logic samples PTBURST# asserted, so it knows more data is being requested. The Add-On keeps WR# asserted, and drives DATA4 onto the DQ bus. PTRDY# is also asserted to complete the current data phase. Clock 8: A s the S5920 samples WR# asserted, it writes DATA4 into the PT Read FIFO. PTRDY# is sampled asserted, which completes the final data transfer and updates the internal FIFO pointers. The S5920 deasserts PTATN#, indicating that the final transfer was performed. No more data is being requested from PCI. The Add-On logic samples PTBURST# deas serted, so it knows that the previous data transfer was the last, and no more data is being requested. The Add-On deasserts WR#, ADR[6:2], SELECT#, BE[3:0]# and DQ. It also deasserts PTRDY#. Note that in a synchronous design, the Add-On logic does not require PTATN# in order to terminate a Pass-Thru read operation, PTBURST# is used for this. Clock 9: As PTATN# and PTBURST# are deasserted, the Pass-Thru access is complete, and the S5920 can accept new Pass-Thru accesses starting on the next clock. The other Pass-Thru signals can also change state (in anticipation of a new transfer). NOTE: W ith prefetch disabled, the performance of Pass-Thru burst reads will be less than optimal. Because of certain issues involving synchronizing signals across clock boundaries (ADCLK -> PCLK), Pass-Thru burst reads will occur only in double and single accesses. For example, a Pass-Thru burst read of five data phases would translate to a burst-read of two DWORDs, another burst-read of two DWORDs followed by a single burst-read with PTATN# being deasserted between each burst packet, losing potentially valuable clock cycles. It is recommended to enable prefetch if maximum performance is desired. Figure 10 also shows a Passive Mode Pass-Thru burst read, but the Add-On logic uses PTRDY# to control the rate at which data is transferred. In many applications, Add-On logic is not fast enough to provide data every ADCLK. In this example, the Add-On interface writes data every other clock cycle. Using PTRDY# to assert Wait-States Clock 0: PCI address information is stored in the PassThru Address Register. The address is recognized as a PCI read of Pass-Thru region 1. Add-On bus signals PTATN#, PTBURST#, PTNUM[1:0], PTWR and PTBE[3:0] will update on the next rising edge of ADCLK. Clock 1: P ass-Thru signals PTATN#, PTBURST#, PTNUM[1:0], PTWR and PTBE[3:0] are driven to indicate what action is required by Add-On logic. These status signals are valid only when PTATN# is active. Add-On logic can decode status signals upon the assertion of PTATN#. PTATN# Asserted. Indicates Pass-Thru access is pending. PTBURST# Asserted. The access has multiple data phases. PTNUM[1:0] 0h.Indicates. the access is to Pass-Thru region 0. PTWR Deasserted. Indicates the access is a read. PTBE[3:0]# D1. Indicates valid bytes for the first data transfer. Clock 2: The Add-On logic has sampled PTATN# and PTBURST# active, indicating that at least two read data transfers are requested by the PCI. The Add-On DS1596 128 AMCC Confidential and Proprietary S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book will start servicing the Burst Read transfer by first reading the Pass-Thru Address via PTADR#. This is an asynchronous read, meaning that the address will appear on DQ after a propagation delay from the assertion of PTADR#. In the event that the address is not required, this cycle and the next could be skipped (as the next clock provides a turn-around cycle). Clock 3: The Add-On logic will latch the Pass-Thru address on the rising edge of this clock. This cycle is also required to avoid contention on the DQ bus. Time must be allowed after PTADR# is deasserted for the DQ outputs to float before Add-On logic attempts to write to the Pass-Thru Read FIFO. Clock 4: T he BE[3:0]#, ADR[6:2], and SELECT# inputs are asserted. WR# and DQ are asserted, indicating that DATA1 is to be written to the PT Read FIFO on the next clock. PTRDY# is asserted, to indicate the completion of the current data phase. Clock 5: A s the S5920 samples WR# asserted, it writes DATA1 into the PT Read FIFO. PTRDY# is sampled asserted, which completes the first data transfer and updates the internal FIFO pointers. The PTBE# outputs are updated to indicate which bytes are valid for the second transfer. The Add-On logic samples PTBURST# asserted, so it knows more data is being requested WR# remains asserted, and DATA2 is driven onto DQ. PTRDY# is also asserted to complete the current data phase. Clock 6: A s the S5920 samples WR# asserted, it writes DATA2 into the PT Read FIFO. PTRDY# is sampled asserted, which completes the second data transfer and updates the internal FIFO pointers. The PTBE# outputs are updated to indicate which bytes are valid for the third transfer. The Add-On logic samples PTBURST# asserted, so it knows more data is being requested. However, it is not ready to transfer data yet, so it deasserts PTRDY# and WR#, and stop driving the DQ bus. The DQ bus could be in tri-state. Clock 7: As the S5920 samples WR# and PTRDY# deasserted, no data was written to the PT Read FIFO and the FIFO pointer was not updated (as the transfer was not signaled complete via a PTRDY#). The AddOn logic is ready to continue the transfer, so it asserts WR# and drives the DQ bus with DATA3. PTRDY# is also asserted to complete the current data phase. Clock 8: A s the S5920 samples WR# asserted, it writes DATA3 into the PT Read FIFO. PTRDY# is sampled asserted, which completes the third data transfer and updates the internal FIFO pointers. The S5920 deasserts PTBURST#, indicating that the previus read was the second to last. The next transfer from the Add-On will be the last. The PTBE# outputs are updated to indicate which bytes are valid for the last transfer. The Add-On logic samples PTBURST# asserted, so it knows more data is being requested. However, it is not ready to transfer data yet, so it deasserts PTRDY# and WR#. The data on the DQ bus is a donÕt care, as the Add-On is not writing during this cycle. The DQ bus could be in tri-state. Clock 9: As the S5920 samples WR# and PTRDY# deasserted, no data was written to the PT Read FIFO and the FIFO pointer was not updated (as the transfer was not signaled complete via a PTRDY#). The AddOn logic samples PTBURST# deasserted and PTATN# asserted, so it knows that the previous data transfer was the last, and no more data is being requested. However, as it inserted a wait state during the previous cycle, it still has one more transfer to complete. As the Add-On logic is ready to complete the transfer, it asserts WR# and drives the DQ bus with DATA4. PTRDY# is also asserted to complete the last data phase. Clock 10: As the S5920 samples WR# asserted, it writes DATA4 into the PT Read FIFO. PTRDY# is sampled asserted, which completes the last data transfer and updates the internal FIFO pointers. The S5920 deasserts PTATN#, indicating that the final transfer was performed. No more data is being requested from PCI. Since the Add-On logic previously sampled PTBURST# deasserted, and transferred the last data, it knows that no more data is being requested. The Add-On deasserts WR#, ADR[6:2], SELECT#, BE[3:0]# and DQ. It also deasserts PTRDY#. Clock 11: A s PTATN# and PTBURST# are deas serted, the Pass-Thru access is complete, and the S5920 can accept new Pass-Thru accesses starting on the next clock. The other Pass-Thru signals can also change state (in anticipation of a new transfer). 8-Bit and 16-Bit Pass-Thru Add-On Bus Interface in Passive Mode The S5920 allows a simple interface to devices with 8bit or 16-bit data buses. Each Pass-Thru region may be defined as 8, 16 or 32 bits, depending on the contents of the boot device which is loaded into the PCI Base Address Configuration Registers during initialization. The result of the initialization is a unique bussize (8/16/32 bits) for each Pass-Thru region. The Pass-Thru Add-On interface internally controls byte lane steering to allow access to the 32-bit Pass-Thru Data FIFO (APTD) from 8-bit or 16-bit Add-On buses. The four DQ data bytes are internally steered depending upon the bus size of the region and the values of the Byte Enables (BE#). Note that this 8-/16-bit internal byte-lane steering is not performed for other Add- AMCC Confidential and Proprietary DS1596 129 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book On operation registers, just the APTD register (ADR = 2Ch). Table 54. Byte Lane Steering for PCI Write (Add-On Read) Byte Enables 3 x x x 0 2 x x 0 1 1 x 0 1 1 0 0 1 1 1 2, PTBE2# is deasserted, and the Add-On may assert PTRDY#, completing the cycle. Table 2 shows how the external Add-On data bus is steered to the Pass-Thru Data Register bytes. This mechanism is determined by the Pass-Thru region bus width defined during initialization. The BYTEn symbols indicate data bytes in the Pass-Thru Data Register. For example, an 8-bit Add-On write with BE1# asserted results in the data on DQ[7:0] being steered into BYTE1 of the APTD register. To write data into the APTD Register, PTBEn# and BEn# must both be asserted. The following describes how APTD writes are controlled: Write BYTE3 if PTBE3# AND BE3# are asserted Write BYTE2 if PTBE2# AND BE2# are asserted Write BYTE1 if PTBE1# AND BE1# are asserted Write BYTE0 if PTBE0# AND BE0# are asserted After each byte is written into the Pass-Thru data registe r, its co rre sp on ding PT BE[3: 0]# ou tpu t is deasserted. This allows Add-On logic to monitor which bytes have been written, and which bytes remain to be written. When all requested bytes have been written (all PTBE[3:0]#s are deasserted), PTRDY# is asserted by the Add-On, completing the access. There are two methods of accessing the Add-On Pass-Thru Address Register (APTA): by asserting the PTADR# pin (and getting the address on DQ after some propagation delay) or by asserting RD#, SELECT, BE[3:0]#’s, and ADR[6:2] = 28h (and getting the address on DQ one cycle later). When using the PTADR# input, all 32 bits of address are driven on DQ, regardless of the state of the DQMODE pin. When accessing APTA via an Add-On operation register access, all 32 bits of address are driven on DQ as long as DQMODE indicates 32 bits. If DQMODE is set for 16 bits, it is necessary to perform two accesses: one with BE[3]# low for the lower 16 bits, and one with BE[3]# high for the upper 16 bits. The Pass-Thru region bus-sizes have no effect on APTA accesses. Figure 11 shows a Pass-Thru write operation for a region defined for an 8-bit Add-On bus interface. As the 8-bit device is connected only to DQ[7:0], the device must access the APTD one byte at a time. A PCI initiator has performed a posted burst-write of two DWORDs to Pass-Thru region zero. Data0 = 08D49A30h and Data1 = AABBCCDDh. All byteenables of the DWORDs were active. Clock 0: The address is recognized as a PCI write to Pass-Thru region 0. The PCI bus write address is stored in the Pass-Thru Address Register. The PCI bus write data is stored in the S5920 write FIFO. AddDS1596 130 APTD Register Write Byte Lane Steering DQ[31:24] BYTE3 BYTE3 BYTE3 BYTE3 DQ[23:16] BYTE2 BYTE2 BYTE2 BYTE3 DQ[15:8] BYTE1 BYTE1 BYTE2 BYTE3 DQ[7:0] BYTE0 BYTE1 BYTE2 BYTE3 Table 55. Byte Lane Steering for PCI Read (Add-On Write) Defined PT Bus Width APTD Register Write Byte Lane Steering BYTE3 32 Bit Data Bus 16 Bit Data Bus 8 Bit Data Bus DQ[31:24] DQ[15:8] DQ[7:0] BYTE2 DQ[23:16] DQ[7:0] DQ[7:0] BYTE1 DQ[15:8] DQ[15:8] DQ[7:0] BYTE0 DQ[7:0] DQ[7:0] DQ[7:0] For Pass-Thru writes (Add-On APTD reads), Add-On logic must read the APTD register one byte or one word at a time (depending on the Add-On bus width). The internal data bus is steered from the correct portion of APTD using the BE[3:0]# inputs. Table 1 shows the byte lane steering mechanism used by the S5920. The BYTEn symbols indicate data bytes in the PassThru Data Register. When a read by the Add-On is performed with a BEn# input asserted, the corresponding PTBEn# output is deasserted. Add-On logic cycles through the byte enables to read the entire APTD Register. Once all data is read (all PTBE[3:0]#s are deasserted), PTRDY# is asserted by the Add-On, completing the access. For Pass-Thru reads (Add-On APTD writes), the bytes requested by the PCI initiator are indicated by the PTBE[3:0]# outputs. Add-On logic uses the PTBE[3:0]# signals to determine which bytes must be written (and which bytes have already been written). For example, a PCI initiator performs a byte Pass-Thru read from an 8-bit Pass-Thru region with PCI BE2# asserted. On the Add-On interface, PTBE2# is asserted, indicating that the PCI initiator requires data on this byte lane. Once the Add-On writes APTD, byte AMCC Confidential and Proprietary S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book On bus signals PTATN#, PTBURST#, PTNUM[1:0], PTWR and PTBE[3:0] will update on the next ADCLK. Clock 1: P ass-Thru signals PTATN#, PTBURST#, PTNUM[1:0], PTWR and PTBE[3:0] are driven to indicate what action is required by Add-On logic. These status signals are valid only when PTATN# is active. Add-On logic can decode status signals upon the assertion of PTATN#. the APTA are output on the DQ bus when PTADR# is asserted. The Add-On must be capable of latching the upper 24 bits (if needed). The Add-On begins reading the APTD Register (asserting SELECT#, ADR[6:2], and RD#). The Add-On logic sees that all bytes are valid (PTBE# = 0h), so starts the read by asserting BE0#, to indicate that BYTE0 of the APTD is to be driven on DQ[7:0] during the next clock cycle. Clock 3: T he Add-On logic latches the Pass-Thru address. RD# and BE0# are sampled by the S5920, so BYTE0 of the APTD is driven on DQ[7:0] and PTBE0# is deasserted. The Add-On asserts RD# and BE1#, thus requesting that BYTE1 of the APTD be driven on the DQ bus during the next cycle. Clock 4: The Add-On logic latches BYTE0. RD# and BE1# are sampled asserted by the S5920, so BYTE1 of the APTD is driven on DQ[7:0] and PTBE1# is deasserted. The Add-On device asserts RD# and BE2#, thus requesting that BYTE2 of the APTD be driven on the DQ bus during the next cycle. PTATN# Asserted. Indicates Pass-Thru access is pending. PTBURST# Asserted. The access has multiple data phases. PTNUM[1:0] 0h. Indicates the access is to Pass-Thru region 0. PTWR Asserted. Indicates the access is a write. PTBE[3:0]# 0h. Indicates valid bytes for the first data transfer. Clock 2: The Add-On sees that a burst-write is being requested by the PCI, so starts by reading the corresponding address via PTADR#. Note that all 32 bits of Figure 69. PCI to Add-On Passive Write to an 8-bit 0 1 2 3 4 5 6 7 8 9 10 11 12 13 ADCLK PTATN# PTBURST# PTNUM[1:0] PTWR PTBE[3:0]# SELECT# ADR[6:2] BE[3:0]# RD# DQ[7:0] DQ[31:8] PTADR# PTRDY# ADD[7:0] 0h 0h 1h 3h 7h 0h 8h Ch Eh Fh 2Ch Eh Dh Bh 7h 7h Bh Dh Eh 30h 9Ah D4h 08h AAh BBh CCh DDh ADD[31:8] AMCC Confidential and Proprietary DS1596 131 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Clock 5: The Add-On logic latches BYTE1. RD# and BE2# are sampled asserted by the S5920, so BYTE2 of the APTD is driven on DQ[7:0] and PTBE2# is deasserted. The Add-On device asserts RD# and BE3#, thus requesting that BYTE3 of the APTD be driven on the DQ bus during the next cycle. PTRDY# is also asserted, indicating that the transfer is complete. Clock 6: The Add-On logic latches BYTE2. RD# and BE3# are sampled asserted by the S5920, so BYTE3 of the APTD is driven on DQ[7:0]. PTRDY# is sampled asserted, so the previous transfer is complete. The PTBE# signals are updated to indicate which bytes are valid for the next transfer (in this case, all bytes are valid for the second DWORD, so PTBE# = 0h). The S5920 deasserts PTBURST#, as it only has one DWORD left to transfer. The Add-On device asserts RD# and BE3#, thus requesting that BYTE3 of the second DWORD in the APTD be driven on the DQ bus during the next cycle. Clock 7: The Add-On logic latches BYTE3 of the first DWORD. RD# and BE3# are sampled asserted by the S5920, so BYTE3 of the second DWORD in the APTD is driven on DQ[7:0] and PTBE3# is deasserted. The Add-On device asserts RD# and BE2#, thus requesting that BYTE2 of the APTD be driven on the DQ bus during the next cycle. Clock 8: The Add-On logic latches BYTE3 of the second DWORD. RD# and BE2# are sampled asserted by the S5920, so BYTE2 of the APTD is driven on DQ[7:0] and PTBE2# is deasserted. The Add-On asserts RD# and BE1#, thus requesting that BYTE1 of the APTD be driven on the DQ bus during the next cycle. Clock 9: The Add-On logic latches BYTE2 of the second DWORD. RD# and BE1# are sampled by the S5920, so BYTE1 of the APTD is driven on DQ[7:0] and PTBE1# is deasserted. The Add-On asserts RD# and BE0#, thus requesting that BYTE0 of the APTD be driven on the DQ bus during the next cycle. PTRDY# is also asserted, indicating that the transfer is complete. As PTBURST# is already deasserted, the AddOn recognizes that this is the last transfer. Clock 10: T he Add-On logic latches BYTE1 of the second DWORD. RD# and BE0# are sampled by the S5920, so BYTE0 of the APTD is driven on DQ[7:0]. PTRDY# is sampled asserted, so the previous transfer is complete. The PTBE# signals are updated to indicate which bytes are valid for the next transfer (in this case, there is no more valid data to transfer, so PTBE = Fh). The S5920 deasserts PTATN#, as it has no data left to transfer. The Add-On device deasserts RD#, BE#, ADR[6:2], SELECT# as the data transfer is complete. AMCC Confidential and Proprietary Clock 11: T he Add-On logic latches BYTE0 of the second DWORD. PTATN# and PTBURST# both deasserted indicate that the Pass-Thru transfer is complete. The PCI can start another access on the next clock cycle. For 16-bit peripheral devices, the byte steering works in the same way. Because the Add-On data bus is 16 bits wide, only two 16-bit cycles are required to access the entire APTD Register. Two byte enables can be asserted during each access. F igure 12 shows a Pass-Thru read operation for a region defined for a 16-bit Add-On bus interface. As the 16-bit device is connected only to DQ[15:0], the device must access the APTD one word at a time. The Add-On must be capable of latching the upper 16 bits of the APTA (if they are needed). The PCI initiator has requested a 32-bit burst read from Pass-Thru region three. All PTBE#s are asserted. Clock 1: T he Add-On begins by reading the APTA register (asserting PTADR#). All 32 bits of the address are driven on the DQ bus. Clock 2: Turn-around cycle, preventing potential bus contention on the DQ bus. Clock 3: The Add-On initiates the write by asserting WR#, SELECT#, BE[3:0]# = “1100”, ADR[6:2] = 2Ch and the low word of the first DWORD to be transferred (D0-LO). Clock 4: The S5920 updates the PTBE#s to indicate that the low word was provided, and that the upper word is still required. The Add-On drives the upper word (D0-HI), and activates the appropriate byte enables, BE# = 0011 The Add-On also asserts PTRDY#, indicating that it is done with the current DWORD, and to advance the FIFO pointer and prepare for the second DWORD. Clock 5: The PTBE#s are updated to indicate that the next DWORD to be transferred requires all bytes. The Add-On drives DQ[15:0] with the lower word of the second DWORD (D1-LO), and the byte-enables indicate the same, BE# = 1100. The Add-On also deasserts PTRDY#. This process continues until the transfer is complete and all words have been written. Endian Conversion Endian conversion can be enabled/disabled for each Pass-Thru Region. It is controlled by bits 6, 14, 22 and 30 of the PTCR. The default endian type for the S5920 is Little Endian. For this reason, the default values in the PTCR are for Little Endian. If Big Endian is selected, the Pass-Thru data and byte-enable interface will be converted to Big Endian type. DS1596 132 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book When the device is programmed for Big Endian translation and a 32-bit data bus, the S5920 will convert as described in Table 3. Active mode is provided to simplify logic requirements when interfacing an application to the Add-On Local bus. Passive mode requires Add-On logic to assert read/write signals and drive or latch data on the DQ bus. Strapping PTMODE low configures the S5920 for Active mode operation. Active mode allows more designer flexibility through programmable features. The following is a brief description of these features. • Pass-Thru address can be driven automatically at the beginning of all transfers or can be skipped altogether if addresses are unneeded by Add-On logic. Programmed or Add-On controlled wait states to delay data transfers automatically or on the fly. Endian Conversion Write FIFO ( Write posting ) Read FIFO ( Prefetch ) PTADR# has been programmed to be output it will go active (low) at this time, and the data presented on the DQ bus is the address for the current transaction. AddOn logic may latch the address value at the rising edge of the clock. Address cycles do not count toward the number of wait states needed to complete data phases. In Active mode, the PTRDY# pin is renamed to PTWAIT#. On cycles after PTWAIT# is sampled low, the state machine is idle. Idle cycles are also not counted as wait states by the S5920. To control the number of wait states on an as-needed basis only, zero wait states should be programmed and PTWAIT# can be driven low when wait states are to be inserted. If PTWAIT# is low when PTATN# is asserted by the S5920, the pending transfer cycle won’t be started until PTWAIT# is driven high. Table 56. Showing Big Endian Conversion for 32-bit Byte# PCI Byte Add-On Byte • • • • 0 1 2 3 D7-D0 D15-D8 D23-D15 D31-D24 D31-D24 D23-D16 D15-D8 D7-D0 Active Operation In Active mode, a data transfer start is signaled on the first clock edge in which PTATN# is sampled low. If Figure 70. PCI to Add-On Passive Read to an 16-bit Add-On Device 0 ADCLK 1 2 3 4 5 6 7 8 9 10 11 12 13 PTATN# PTBURST# PTNUM[1:0] 3h PTWR PTBE[3:0]# 0h 3h 0h 3h 0h 3h Fh SELECT# ADR[6:2] 2Ch Ch 3h Ch 3h Ch 3h BE[3:0]# WR# ADDR [31:16] ADDR [15:0] D1 LOW D1 HIGH D2 LOW D2 HIGH D3 LOW D3 HIGH DQ[31:16] DQ[15:0] PTADR# PTRDY# AMCC Confidential and Proprietary DS1596 133 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Figure 71. Active mode PCI Read (Zero Programmed Wait States) with PTADR# Table 57. Big Endian conversion for a 16-bit bus. The S5920 drives D[15:0] only Transfer Byte # PCI Byte Lane Add-On Bus Byte Lane 1 ADCLK PTATN# PTBURST# PTNUM[1:0] PTWR PTBE[3:0] DXFR# DQ[31:0] PTWAIT# PTADDR DATA 0h Fh 01b 2 3 4 5 6 7 1st XFER 1st XFER 2nd XFER 2nd XFER 0 1 2 3 D7-D0 D15-D8 D23-D16 D31-D24 D15-D8 D7-D0 D15-D8 D7-D0 Table 58. Big Endian conversion for an 8-bit bus. The S5920 drives D[7:0] only Transfer Byte # PCI Byte Lane Add-On Bus Byte Lane PTADR# Figure 72. Active Mode PCI Read without PTADR# 1 2 3 4 5 6 7 1st XFER 2nd XFER 3rd XFER 4th XFER 0 1 2 3 D7-D0 D15-D8 D23-D16 D31-D24 D7-D0 D7-D0 D7-D0 D7-D0 ADCLK PTATN# PTBURST# PTNUM[1:0] 01b In Active mode, wait states can also be programmed. This enables easier interfacing to slow Add-On logic whichcannot transferdataatthe fullADCLKspeed.The S5920 inserts a turnaround cycle after the address phase for PCI Read cycles. If one or more wait states have been programmed, the turnaround cycle is considered the first wait state of the first data phase of that transaction. For all Active mode transfers, the DXFR# signal is used by Add-On logic as the data transfer signal. Data must be latched at the rising edge of ADCLK when DXFR# is asserted for a PCI write. Conversely, for PCI Reads, the rising edge of ADCLK when DXFR# is asserted can be used to increment to the next data field. PTWR PTBE[3:0] DXFR# DQ[31:0] PTWAIT# PTADR# DATA 0h Fh The PTADR# signal is controlled by the most significant bit of every region control field in the PTCR register. If this bit is zero then the PTADR# pin is not driven at the start of an Active mode transfer. If this bit is set to one, the PTADR# pin will be enabled and driven active (low) for one and only one clock after PTATN# was sampled active provided PTWAIT# was also sampled high. AMCC Confidential and Proprietary DS1596 134 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Figure 73. Active Mode PCI Write without PTADR# 1 ADCLK PTATN# PTBURST# PTNUM[1:0] PTWR PTBE[3:0] DXFR# DQ[31:0] PTWAIT# PTADR# DATA 0h Fh 01b 2 3 4 5 6 7 is not considered active until PTATN# is low and PTWAIT# is high) regardless of the transfer being a read or a write. Figure 13b shows a PCI read cycle with PTADR# disabled. Figure 13c shows a Pass-Thru write cycle with PTADR# disabled. Active mode Programmable Wait States Bits 0,1,2 of the PTCR register control this feature. Wait States are programmed on a per region basis. For example: region one can be set for zero wait states while other regions may have multiple wait states programmed. Wait state options are 0,1,2,...7 wait states. The S5920 will always count N wait states (N=0,1,..7) before completing the current data phase. Figures 17, 18 and 19 show Pass-Thru transfers with programmed wait states. PTRDY#/PTWAIT# When PTADR# is active (low), the S5920 will drive the DQ[31:0] bus with the 32-bit PCI address regardless of the PTMODE pin. To avoid contention on the DQ[31:0] bus during PCI read cycles, the S5920 incorporates a turnaround cycle before starting to drive the data (DXFR# assertion). This is needed only when PTADR# is enabled and when zero wait states are programmed during a Pass-Thru read cycle. The cycle immediately following the address cycle will be a turnaround cycle as shown in Figure 13a. If PTADR# is disabled, the DXFR# output will be driven one clock cycle after PTATN# is valid (PTATN# In Active mode, the PTRDY#/PTWAIT# pin takes the PTWAIT# function, which is the opposite function of this pin when configured for passive mode. That is, if the part is configured to operate in Active mode, PTWAIT# asserted low means the Add-On wishes to insert wait states. Add-On peripherals are allowed to insert wait state cycles at any time during an Active mode transfer. When PTWAIT# has been sampled low, the S5920 will tri-state its DQ[31:0] bus in order to allow other AddOn devices to use the bus without contention. Figure 74. Active Mode PCI Write with Add-On Initiated Wait States Using PTWAIT# 0 1 2 3 4 5 6 7 8 9 10 11 12 13 ADCLK PTATN# PTBURST# PTNUM[1:0] PTWR PTBE[3:0]# DXFR# DQ[31:0] PTADR# PTWAIT# PTADDR 01b 0h Fh DATA AMCC Confidential and Proprietary DS1596 135 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Figure 75. Active Mode 32-Bit PCI Write 1 2 3 4 5 6 are delayed by an external device controlling PTWAIT#. Clock 1: The S5920 drives PTATN# active (low), indicating the start of a PCI to Add-On data transfer. PTBE[3:0] and PTNUM[1:0] are driven to their appropriate values for this transfer. PTWR is driven high indicating a PCI write. ADCLK PTATN# PTBURST# PTNUM[1:0] PTWR PTBE[3:0] DXFR# DQ[31:0] PTWAIT# PTADR# PTADDR 01b Clock 2: T his is a wait state since PTWAIT# was active (low) at the rising edge of clock 2. Clock 3: PTWAIT# was inactive (high) at the rising edge of clock 3 so this cycle is the address phase: PTADR# is driven active (low) and the address value for the current transaction is driven onto the DQ bus. Clock 4: PTADR# was active (low) at the rising edge of this clock so the Add-On device must latch the PCI address on the rising edge of this clock. The S5920 treats this cycle as a wait state since PTWAIT# was active (low) at the rising edge of clock 4. Clock 5: T his is a wait state since PTWAIT# was active (low) at the rising edge of clock 5. 0h Fh DATA The address phase of a Pass-Thru consists of the cycles from PTATN# asserted through PTADR# asserted (if PTADR# has been programmed to be disabled, there is no address phase). If PTWAIT# is asserted before the address phase, the address phase is delayed. The address phase will occur during the cycle after the clock edge that PTWAIT# is sampled high and PTATN# is sampled low. The data phase(s) of a Pass-Thru consists of all the cycles after the (possibly nonexistent) address phase. During data phases, a wait is incurred during the cycle after PTWAIT# is sampled asserted. Note: If PTWAIT# is activated in order to access other registers internal to the S5920, the user is responsible for inserting any needed turnaround cycles in order to avoid bus contention on the DQ bus. DXFR# Clock 6: PTWAIT# was inactive (high) at the rising edge of clock 6, so DXFR# is driven active (low) indicating a data transfer. PTATN# is driven inactive (high) indicating the Pass-Thru access is complete. Clock 7: DXFR# was active (low) at the rising edge of this clock so the Add-On device must latch the PCI data on the rising edge of this clock. PTBE# is driven to Fh indicating all 4 bytes have been accessed. PTNUM and PTWR may change state since the PassThru access is complete. Clock 8: PTBE# may change state. Figure 15 shows a single data phase 32-bit Active mode PCI Write with PTADR# enabled. Active mode Burst cycles DXFR# is a signal that is active during the cycles that a data transfer may take place. It is intended to be used to control strobes (e.g., write enable, read enable), and can be a flag for incrementing to the next address during a burst. If wait states have been programmed, DXFR# will not go active until after all wait states have been executed. Note that asserting PTWAIT# to insert Add-On initiated wait states causes temporary suspension of the internal programmed wait state counter. Active Mode Figures and Descriptions PTBURST# signifies to the Add-On device that the current transfer will be contain more than one data phase. The Add-On device detects the end of a burst when the S5920 deasserts PTBURST#. During an Active mode PCI burst read, PTBURST# is deasserted when there is one more data word left to transfer. During an Active mode PCI burst write, PTBURST# deasserted indicates that after the current data word is transferred, there will be one data word left to transfer. Figure 16 shows an Active mode PCI Burst Write with 0 programmed wait states. The Add-On device controlling PTWAIT# asserts wait states in the figure on an as-needed basis. PTADR# has been programmed to be disabled. Figure 14 shows a programmed zero wait state transfer in which the cycle start and the cycle completion AMCC Confidential and Proprietary DS1596 136 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Figure 76. Active Mode 32-Bit PCI Write w/PTWAIT# 0 1 2 3 4 5 6 7 8 9 10 11 12 13 ADCLK PTATN# PTBURST# PTNUM[1:0] PTWR PTBE[3:0]# DXFR# DQ[31:0] PTADR# PTWAIT# 01b Data1 Data2 Data3 Data4 Data5 Fh DATA1 DATA2 DATA3 DATA4 DATA5 Clock by Clock description of Figure 16 Clock 1: The S5920 drives PTATN# and PTBURST# active (low), indicating the start of a PCI to Add-On data transfer with more than one data cycle. PTBE[3:0] and PTNUM[1:0] are driven to their appropriate values for this transfer. PTWR is driven high indicating a Pass-Thru write. Clock 2: S ince this region does not have PTADR# enabled as an output and PTWAIT# is high at the rising edge of clock 2, the first data transfer is indicated by driving DXFR# low and the data on the data bus DQ[31:0] . Clock 3: D XFR# is sampled active by the Add-On device which indicates that the Add-On device must latch the first data word at the rising edge of this clock. Valid data is determined by decoding the PTBE[3:0]# lines. The Add-On device drives PTWAIT# active (low) requesting a wait state on the next cycle. Clock 4: DXFR# is sampled active (low) by the AddOn device which indicates that the Add-On device must latch the second data word at this clock edge. The S5920 tri-states its output bus since PTWAIT# was inactive (high) at the rising edge of clock 4. Additionally, the S5920 deasserts DXFR# indicating that no data transfer will occur on the next clock edge (this is because this cycle is a wait state since PTWAIT# was active (low) at the rising edge of clock 4. The Add-On device deasserts PTWAIT# indicating no wait state on the next clock. Clock 5: N o data transfer takes place at the rising edge of clock 5 since the previous cycle was an AddOn initiated wait state (because PTWAIT# was active (low) at the rising edge of clock 4). The S5920 asserts DXFR# and drives the third data onto the DQ bus since PTWAIT# was inactive (high) at the rising edge of clock 5. The Add-On device drives PTWAIT# active (low) requesting a wait state on the next cycle. Clock 6: D XFR# is sampled active by the Add-On device which indicates that the Add-On device must latch the third data word at the rising edge of this clock. The S5920 drives DXFR# inactive and tri-states the DQ bus since PTWAIT# was active (low) at the rising edge of clock 6. The Add-On keeps PTWAIT# asserted indicating it wants to add a wait state on the next cycle. Clock 7: No data transfer takes place on the rising edge of this clock since the previous cycle was an Add-On initiated wait state (because PTWAIT# was active (low) at the rising edge of clock 6). Clock 8: No data transfer takes place on the rising edge of this clock since the previous cycle was an Add-On-initiated wait state (because PTWAIT# was active (low) at the rising edge of clock 7). DXFR# is driven active (low) and the fourth data is driven onto the DQ bus since PTWAIT# was inactive (high) at the rising edge of clock 8. PTBURST# is driven inactive (high) indicating that after this data word is transferred, there is only one data word left to transfer. AMCC Confidential and Proprietary DS1596 137 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Clock 9: DXFR# is sampled active (low) by the AddOn device which indicates that the Add-On device must latch the fourth data word at the rising edge of this clock. PTATN# is driven inactive (high) indicating that this will be the last data phase. Clock 10: DXFR# is sampled active (low) by the AddOn device which indicates that the Add-On device must latch the fifth data word at the rising edge of this clock. Clock 11: PTBE# may change state. Figure 17 shows a Pass-Thru Burst Write data transfer in which the S5920 has been programmed to strobe data using a one-wait state delay. The Add-On device leaves PTWAIT# inactive (high) for all time. Clock by Clock description of Figure 17 DXFR# is deasserted since the access is complete. PTBE# is driven to Fh indicating all 4 bytes have been accessed. PTNUM and PTWR may change state since the access is complete. Clock 1: The S5920 drives PTATN# and PTBURST# active (low) indicating the start of a PCI to Add-On data transfer with more than one data cycle. PTBE[3:0] and PTNUM[1:0] are driven to their appropriate values for this transfer. PTWR is driven high indicating a Pass-Thru write. Figure 77. Active Mode PCI Write Showing a One Wait State Programmed Delay 0 ADCLK PTATN# PTBURST# PTNUM[1:0] PTWR PTBE[3:0]# DXFR# DQ[31:0] PTADR# PTWAIT# PTADDR DATA1 DATA2 DATA3 DATA4 Data1 Data2 Data3 Data4 Fh 1h 1 2 3 4 5 6 7 8 9 10 11 12 13 Figure 78. 16-Bit Active Mode PCI Read w/ Programmed Wait States 0 ADCLK PTATN# PTBURST# PTNUM[1:0] PTWR PTBE[3:0]# DXFR# DQ[15:0] PTADR# PTWAIT# PTADDR 1 2 3 4 5 6 7 8 9 10 11 12 13 14 1h Ch 3h Fh LOW HIGH AMCC Confidential and Proprietary DS1596 138 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Clock 2: S ince this region does have PTADR# enabled as an output, it is driven active (low) and the PCI address for the current transaction is presented on the DQ[31:0] bus. Clock 3: T he Add-On device must latch the PCI address at the rising edge of this clock. Clock 4: DXFR# is asserted low indicating that data will be transferred on the next rising clock edge (clock 5). Data1 is driven onto the DQ[31:0] bus. Clock 5: The Add-On device must latch the first data word at the rising edge of this clock. Valid data is determined by decoding the PTBE[3:0]# lines. Clock 6: DXFR# is asserted indicating that data will be transferred on the next rising clock edge (clock 7). DATA2 is driven onto the DQ[31:0] bus. Clock 7: The Add-On side device latches the second data word (DATA2) at the rising edge of this clock. Clock 8: DXFR# is asserted indicating that data will be transferred on the next rising clock edge (clock 9). DATA3 is driven onto the DQ[31:0] bus. PTBURST# is driven inactive indicating that after this data word is transferred, there is only one data word left to transfer. Clock 9: T he Add-On side device latches the third data word (DATA3) at this clock edge. Clock 10: DXFR# is asserted indicating that data will be transferred on the next rising clock edge (clock 11). DATA4 is driven onto the DQ[31:0] bus. PTATN# is deasserted indicating that this will be the last data phase. Clock 11: T he final data word (DATA4) must be latched by the Add-On device at the rising edge of this clock. PTBE# is driven to Fh indicating all 4 bytes have been accessed. PTNUM and PTWR may change state since the access is complete. Clock 12: PTBE# may change state. Active Mode with 16/8-bit data buses When the S5920 is programmed in Active mode and 16-bit, the DXFR# output will strobe twice for every PCI 32-bit word that has been read/written. Each DXFR# assertion signifies that a 16-bit word has been transferred to the Add-On side. The first DXFR# completion will be for the least significant 16-bit word of a 32-bit word (“LOW” in Figures 18, 9-20 and 9-21), while the second transfer will be for the most significant 16-bit word (“HIGH” in Figures 18, 9-20 and 921). If the current PCI access has only 2 bytes valid (PCI BE[3:0]# encoding of Ch or 3h instead of 0h), then the S5920 will still assert a 2 cycle completion but one of them will not contain valid data (PTBE[3:0]#=Fh). If the programmed wait states for the current Pass-Thru region is not zero, then the S5920 will insert the programmed wait states before the “LOW” data word and also between the “LOW” and “HIGH” data words. Figure 18 shows a PCI read to a 16-bit Add-On region with two programmed wait states. Note that a PCI read to an 8-bit Add-On would be the same as Figure 18 except that there would be 4 data transfers (one for each byte) vice 2. As in Passive mode, in Active mode, the word read/ write order is determined by the Endian conversion programmed into the S5920. Figure 79. Active Mode PCI Read w/ Programmed Wait States 0 ADCLK PTATN# PTBURST# PTNUM[1:0] PTWR PTBE[3:0]# DXFR# DQ[31:0] PTADR# PTWAIT# PTADDR 1 2 3 4 5 6 7 8 9 10 11 12 13 14 1h Data1 Data2 Data3 Fh DATA1 DATA2 DATA3 AMCC Confidential and Proprietary DS1596 139 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Figure 80. Active Mode PCI Read 1 2 3 4 5 6 Figure 81. Active Mode PCI Write 1 2 3 4 5 6 ADCLK PTATN# PTBURST# PTNUM[1:0] PTWR PTBE[3:0] DXFR# DQ[15:0] PTWAIT# PTADR# ADCLK PTATN# PTBURST# 1h PTNUM[1:0] PTWR 1h Ch 3h Fh PTBE[3:0] DXFR# Ch 3h Fh LOW HIGH DQ[15:0] PTWAIT# PTADR# LOW HIGH Figure 20 shows a Pass-Thru write cycle with 0 wait states for a 16-bit region. The PTBE[3:0]# signals are used by the Add-On device to determine validity of the current word cycle and also, which word of a long word is currently being driven by the S5920. PTBE[3:0]# encoding of Ch indicates the least significant 16-bit portion of the 32-bit PCI data word is on the DQ[15:0] bus. PTBE[3:0]# encoding of 3h indicates the most significant 16-bit portion of the 32-bit PCI data word is on the DQ[15:0] bus. PTADR# is shown as disabled in Figure 20. Figure 21 shows a Pass-Thru read cycle with 0 wait states for a 16-bit region. PTADR# is disabled. If the Add-On bus size is 8 bits, then the S5920 will assert DXFR# 4 times for each 32-bit PCI word The first completion is for byte 0, the second is for byte 1, the third is for byte 3, and the fourth DXFR# assertion is for byte 4 of a 32-bit word. If the current PCI access has less than four bytes valid (PCI BE[3:0]# encoding is not 0h), then the S5920 will still assert a 4-cycle completion but one or more of them will not contain valid data (PTBE[3:0]# = Fh). As in Passive mode, in Active mode, the word read/ write order is determined by the Endian conversion programmed into the S5920. Figure 22 shows a Pass-Thru write cycle with 0 wait states for an 8-bit region. The PTBE[3:0]# signals are used by the Add-On device to determine validity of the current byte cycle and also, which byte of a long word is currently being driven by the S5920. PTADR# is enabled as an output. CONFIGURATION The S5920 Pass-Thru interface utilizes four Base Address Registers (BADR1:4). Each Base Address Register corresponds to a Pass-Thru region. The contents of these registers during initialization determine the characteristics of that particular Pass-Thru region. Each region can be mapped to memory or I/O space. Memory mapped devices can, optionally, be mapped below 1 Mbyte and can be identified as prefetchable. Both memory and I/O regions can be configured as 8, 16 or 32 bits wide. Base Address Registers are loaded during initialization from the external non-volatile boot device. Without an external boot device, the default value for the BADR registers is zero (region disabled). The Base Address Registers are the only registers that define Pass-Thru operation. Consequently, the Pass-Thru interface cannot be used without an external non-volatile boot device. AMCC Confidential and Proprietary DS1596 140 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book S5920 Base Address Register Definition Certain bits in the Base Address Register have specific functions: D0 PCI BIOS will not allocate the I/O space and will probably disable the region. Figure 82. 8-Bit Active Mode PCI Write 1 ADCLK PTATN# PTBURST# PTNUM[1:0] PTWR PTBE[3:0] DXFR# Memory or I/O mapping. If this bit is clear, the region should be memory mapped. If this bit is set, the region should be I/O mapped. Location of a memory region. These bits request that the region be mapped in a particular part of memory. These bit definitions are only used for memory mapped regions. D2 D1 Location 2 3 4 5 6 7 D2:1 1h 0 0 1 1 D3 0 1 0 1 Anywhere in 32-bit memory space Below 1 Mbyte in memory space (Real Mode address space) Anywhere in 64-bit memory space (not valid for the S5920) Reserved Eh Dh Bh 7h Fh DQ[7:0] PTWAIT# PTADR# PTADDR Byte0 Byte1 Byte2 Byte3 Prefetchable. For memory mapped regions, the region can be defined as cacheable. If set, the region is cacheable. If this bit is clear, the region is not. Pass-Thru region bus width. These two bits are used by the S5920 to define the data bus width for a Pass-Thru region. Regardless of the programming of other bits in the BADR register, if D31:30 are zeros, the Pass-Thru region is disabled. D31 D30 Add-On Bus Width Creating a Pass-Thru Region D31:30 Section 3.11 describes the values that must be programmed into the non-volatile boot device to request various block sizes and characteristics for Pass-Thru regions. After reset, the S5920 downloads the con tents of the boot device locations 54h, 58h, 5Ch, and 60h into “masks” for the corresponding Base Address Registers. The following are some examples for various Pass-Thru region definitions: NV Memory Contents Pass-Thru Region Definition 0 0 1 1 0 1 0 1 Region disabled 8 bits 16 bits 54h = BFFFF002h Pass-Thru region 1 is a 4 Kbyte region, mapped below 1 Mbyte in memory space with a 16-bit Add-On data bus. This memory region is not cacheable. Pass-Thru region 2 is disabled. (D31:30 = 00.) Pass-Thru region 3 is a 32-bit, 128 byte I/O-mapped region. Pass-Thru region 4 is disabled. 58h = 3xxxxxxxh 32 bits 5Ch = FFFFFF81h 60h = 00000000h BADR1:4 bits D31:30 are used only by the S5920. When the host reads the Base Address Registers during configuration cycles, they always return the same value as D29. If D29 is zero, D31:30 return zero, indicating the region is disabled. If D29 is one, D[31:30] return one. This operation limits each Pass-Thru region to a maximum size of 512 Mbytes of memory. For I/O mapped regions, the PCI specification allows no more than 256 bytes per region. The S5920 allows larger regions to be requested by the Add-On, but a During the PCI bus configuration, the host CPU writes all ones to each Base Address Register, and then reads the contents of the registers back. The mask downloaded from the boot device determines which bits are read back as zeros and which are read back as ones. The number of zeros read back indicates the amount of memory or I/O space a particular S5920 Pass-Thru region is requesting. DS1596 141 AMCC Confidential and Proprietary S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book After the host reads all Base Address Registers in the system (as every PCI device implements from one to six), the PCI BIOS allocates memory and I/O space to each Base Address region. The host then writes the start address of each region back into the Base Address Registers. The start address of a region is always an integer multiple of the region size. For example, a 64-Kbyte memory region is always mapped to begin on a 64K boundary in memory. It is important to note that no PCI device can be absolutely located in system memory or I/O space. All mapping is determined by the system, not the application. PCI or Add-On Operation registers PTCR or APTCR provide additional configuration control for each region. Accessing a Pass-Thru Region location in the external nvRAM which will contain these custom programmed bits. These features, and their corresponding bit at location 45h, are described as follows: LOC_45(h) Descripton Default b=0 b=1 b=2 b = (7:3) Target Latency Enb Retry Flush Enb Write FIFO Mode Reserved 1 0 0 x After the system is finished defining all Base Address Regions within a system, each Base Address Register contains a physical address. The application software must now find the location in memory or I/O space of its hardware. PCI systems provide BIOS or operating system function calls for application software to find particular devices on the PCI bus based on Vendor ID and Device ID values. This allows application software to access the device’s Configuration Registers. The Base Address Register values in the S5920’s Configuration Space may then be read and stored for use by the program to access application hardware. The value in the Base Address Registers is the physical address of the first location of that Pass-Thru region. Some processor architectures allow this address to be used directly to access the PCI device. For Intel Architecture systems, the physical address must be changed into a Segment/Offset combination. For Real Mode operation in an Intel Architecture system (device mapped below 1 Mbyte in memory), creating a Segment/Offset pair is relatively simple. To calculate a physical address, the CPU shifts the segment register 4 bits to the left and adds the offset (resulting in a 20 bit physical address). The value in the Base Address Register must be read and shifted 4 bits to the right. This is the segment value and should be stored in one of the Segment registers. An offset of zero (stored in SI, DI or another offset register) accesses the first location in the Pass-Thru region. Special Programming Features Target Latency describes the number of cycles that a target device may respond to a PCI data transfer request. The PCI 2.1 specification indicates that the target device has 16 clocks to respond to an initial request (from the assertion of FRAME#), and 8 clocks from each subsequent data phase. If the target is not capable of asserting TRDY# within these time frames, it must assert a STOP#, thus initiating a disconnect. The Target Latency programmed bit allows the user to disable the generation of disconnects in the event of a slow Add-On device. If Target Latency Enb is low, target latency is ignored. In this case, the S5920 will never issue a retry/disconnect in the event of a slow add-on device. Instead, TRDY# wait states will be asserted. This might be useful for an embedded system, where the S5920 can take up as many clock cycles as necessary to complete a transfer. This programmable bit is only provided for flexibility and most users should leave this bit set to 1. If Target Latency Enb is high, the device will be PCI 2.1 compliant with respect to Target Latency. Retry Flush Enb indicates to the Pass-Thru whether to hold prefetched data following a disconnect, or to allow the data to be flushed out during the next PCI read access. If low, the data will be held in the PT read FIFO until the initiator comes back to read it out. All subsequent PCI accesses to the S5920 from a device other than the one who initiated the read will be acknowledged with a retry. If the master never returns for the data, the Pass-Thru function will be hung. Even though the PCI 2.1 Specification does not require a master to return for data following a disconnect, it is unlikely that a master will terminate a read transfer until all data has been collected. If Retry Flush Enb is high, the data will be flushed from the FIFO if a subsequent PCI read access is not to the same address. If the original master received a retry DS1596 142 A few additional features have been provided to the user which will allow for optimal “tuning” of their system. As these are not features that will be changed “on the fly”, they have been included as part of the nvRAM boot-up sequence. nvRAM address 45h is a memory AMCC Confidential and Proprietary S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book (disconnect, but with no data transfer), the read data is held in the FIFO until the master comes back for it. In this case, the Retry Flush Enb has no effect. The PCI 2.1 Specification states that the master must come back if it receives a retry. Write FIFO Mode indicates what to do in the event that a full FIFO is detected during a PCI write transfer. If low, the S5920 will perform an immediate disconnect, thus freeing up the PCI bus for other transfers. The initiator will have to come back to complete the data transfer, after which time the FIFO should no longer be full. If Write FIFO Mode is high, the S5920 will deassert TRDY#, and allow for either another FIFO location to become available (as the Add-On has read a DWORD), or wait for the Target Latency to expire (8 clocks from previous data phase), thus initiating a disconnect. This will allow for the Add-on device to “catch-up” without losing the burst. AMCC Confidential and Proprietary DS1596 143 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book ABSOLUTE MAXIMUM STRESS RATINGS Table 1 lists the absolute maximum S5920 device stress ratings. Stresses beyond those listed may cause permanent damage to the device. These are stress ratings only; operation of the device at these or any other conditions beyond those indicated in the Operating Characteristics section of this specification is not implied. Table 59. Absolute Maximum Stress Ratings Parameter Min Max Units Storage Temperature Supply Voltage (V DD ) Input Pin Voltage -55 - 0.3 - 0.5 125 7.0 GND + 5.0 °C Volts Volts S5920 Operating Conditions Table 2 lists the S5920 operating conditions. These parameters are guaranteed by device characterization, but not device tested. All values are maximum guaranteed values. Table 60. Operating Conditions Symbol Parameter Min Max Units Test Conditions Notes VCC ICC ICC TA 0JA PDD Supply Voltage Supply Current (Static) Supply Current (Dynamic) Operating Temperature Thermal Resistance Power Dissipation 4.75 - 49 mA - 197 mA 0 5.25 VCC = 5.25 VCC = 5.25 70 - TBD V V V °C °C/W W V CC = 5.25 Volts 0 TBD AMCC Confidential and Proprietary DS1596 144 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book PCI Signal DC Characteristics The following table summarizes the DC characteristic parameters for all PCI signals listed below as they apply to the S5920. AD[31:0] (t/s), PAR (t/s), C/BE[3:0]# (in), FRAME# (in), IRDY# (in), TRDY# (s/t/s), STOP# (s/t/s), LOCK# (in), IDSEL (in), DEVSEL# (s/t/s), PERR# (s/t/s), SERR# (o/d), INTA# (o/d), RST# (in), CLK (in). Table 61. PCI Signal DC Characteristics (V CC = 5.0V 5%, 0 0 C to 70 0 C, 50 pF load on outputs) Symbol Parameter Min Max Units Test Conditions Notes VIH VIL LIH LIL VOH VOL CIN CCLK CIDSEL Notes: Input High Voltage Input Low Voltage Input High Leakage Current Input Low Leakage Current Output High Voltage Output Low Voltage Input Pin Capacitance CLK Pin Capacitance IDSEL Pin Capacitance 2.0 -0.5 2.4 5 - 0.8 70 -70 0.55 10 12 8 Volts Volts uA uA Volts Volts pF pF pF VIN = 2.7 VIN = 0.5 IOUT = -2 mA IOUT = 3 mA, 6 mA 1 1 2 2 3 4 1. Recommended values for all PCI signals except CLK to be Vih = 2.4 Min and Vil = .4 Max. 2. Input leakage applies to all inputs and bi-directional buffers. 3. PCI bus signals without pull-up resistors will provide the 3 mA output current. Signals which require a pull-up resistor will provide 6 mA output current. 4. The PCI specification limits all PCI inputs not located on the motherboard to 10 pF (the clock is allowed to be 12 pF). AMCC Confidential and Proprietary DS1596 145 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Add-On Signal DC Characteristics The following table summarizes the Add-On DC characteristic parameters for the Add-On signals listed below as they apply to the S5920. All Add-On signal outputs listed are capable of sinking or sourcing 4 mA with the exception of BPCLK which can sink or source 8 mA. DQ[31:0] (t/s), ADR[6:2] (in), BE[3:0]# (in), SELECT# (in), RD# (in), WR# (in), PTATN# (out), PTBE[3:0]# (out), PTNUM[1:0] (out), PTWR (out), PTBURST# (out), PTADR# (I/O), PTRDY# (in), PTMODE (in), DXFR# (out), SYSRST# (out), IRQ# (out), ADDINT# (in), BPCLK (out), ADCLK (in), DQMODE (in), MDMODE (in), MD[7:0] (I/O), LOAD# (in). Table 62. Add-On Operating Characteristics (V CC = 5.0V 5%, 0 0 C to 70 0 C, 50 pF load on outputs) Symbol Parameter Min Max Units Test Conditions Notes VIH VIL LIH LIL VOH VOL Input High Voltage Input Low Voltage Input High Leakage Current Input Low Leakage Current Output High Voltage Output Low Voltage 2.0 -0.5 -10 -10 3.5 - 0.8 +10 +10 0.4 Volts Volts uA uA Volts Volts VIN = 2.7 VIN = 0.5 nvRAM Memory Interface Signals SDA 8 mA (bi-directional-o/d) SCL 8 mA (o/d) AMCC Confidential and Proprietary DS1596 146 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book TIMING SPECIFICATION PCI Clock Specification Table 5 summarizes the A. C. characteristics for the PCI bus signals as they apply to the S5920. The figures after Table 5 visually indicate the timing relationships. Table 63. Functional Operation Range (VCC = 5.0V +/- 5%, 0O C to 70 OC, 50 pF load on outputs for MAX, 0 pF load for MIN) Symbol Parameter Min Max Units Notes Tcyc t1 t2 t3 t4 t5 t6 t7 t8 t9 Notes: Clock Time CLK High Time CLK Low Time Rise Time (0.8V to 2.0V) Fall Time (2.0V to 0.8V) CLK to Signal Valid Delay (Bused Signals) CLK to Signal Valid Delay (Point-to-Point Signals) Float to Active Delay Active to Float Delay Rising Edge Setup Hold from PCI Clock Rising Edge 30 11 11 1 1 2 2 2 28 7 0 4 4 11 12 ns ns ns V/ns V/ns ns ns ns ns ns 1 1 1,2 3 3 4 4 1. Rise and fall times are specified in terms of the edge rate measured in V/ns. This slew rate is met across the minimum peak-to-peak portion of the clock waveform as shown in Figure 1. 2. Minimum times are evaluated with 0 pF equivalent load; maximum times are evaluated with 50 pF equivalent load. 3. For purposes of Active/Float timing measurements, the Hi-Z or 'off" state is defined to be when the total current delivered through the component pin is less than or equal to the leakage current specification. 4. See the timing measurement conditions in Figure 3. Figure 83. PCI Clock Timing t1 2.0 0.8 2.0 0.8 t3 VIH2 t2 TCL 2.0 0.8 2.0 0.8 t4 AMCC Confidential and Proprietary DS1596 147 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Figures 2 and 3 define the conditions under which timing measurements are made. The user designs must guarantee that minimum timings are met with maximum clock skew rate (fastest edge) and voltage swing. Figure 84. PCI Signal Output Timing PCI CLK t5 1.5 OUTPUT DELAY 1.5 TRI-STATE OUTPUT t6 1.5 1.5 t7 Figure 85. PCI Signal Input Timing PCI CLK t8 t9 INPUT INPUTS VALID Add-On Signal Timings AMCC Confidential and Proprietary DS1596 148 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Table 6 summarizes the A. C. characteristics for the Add-On bus signals as they apply to the S5920. The figures after Table 6 visually indicate the timing relationships. Table 64. Add-On Timings, Functional Operation Range (VCC = 5.0 V ± 5%, 0°C to 7 0°C, 50 pF load on outputs for MAX, 0 pF load for MIN) Symbol Parameter Min Max Units Notes TACL t10 t11 t12 t13 t14 t15 t16 t17a t17b t18 t19 t20 t21 t22 t23 t24 t25 t26 t27 t28 t29 t30 t31 t32 t33 ADCLK Cycle Time ADCLK High Time ADCLK Low Time ADCLK Rise Time (0.8V to 2.0V) ADCLK Fall Time (2.0V to 0.8V) PCICLK to BPCLK Delay, Rising PCICLK to BPCLK Delay, Falling PTADR# Low to DQ[31:0] Output Valid PTADR# High to DQ[31:0] Output Hold PTADR# High to DQ[31:0] Output Float PTATN# Valid from ADCLK Rising Edge PTATN# Hold from ADCLK Rising Edge PTBURST# Valid from ADCLK Rising Edge PTBURST# Hold from ADCLK Rising Edge PTNUM[1:0] Valid from ADCLK Rising Edge PTNUM[1:0] Hold from ADCLK Rising Edge PTWR Valid from ADCLK Rising Edge PTWR Hold from ADCLK Rising Edge PTBE[3:0]# Valid from ADCLK Rising Edge PTBE[3:0]#Hold from ADCLK Rising Edge PTWAIT# Setup to ADCLK Rising Edge PTWAIT# Hold from ADCLK Rising Edge SELECT# Setup to ADCLK Rising Edge SELECT# Hold from ADCLK Rising Edge ADR[6:2] Setup to ADCLK Rising Edge ADR[6:2] Hold from ADCLK Rising Edge 25 10 10 2.5 2.5 5.3 6.2 13.1 2 11.9 13.5 4 13 4 14.4 4 13.1 4 14.4 3 11 1 8.9 1 9.3 1 ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns 1 1 1 AMCC Confidential and Proprietary DS1596 149 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Table 64. Add-On Timings, Functional Operation Range (VCC = 5.0 V ± 5%, 0°C to 7 0°C, 50 pF load on outputs for MAX, 0 pF load for MIN) (Continued) Symbol Parameter Min Max Units Notes t34 t35 t36 t37 t38 t39 t40 t41 t42 t43 t44 Notes: BE[3:0]# Setup to ADCLK Rising Edge BE[3:0]# Hold from ADCLK Rising Edge RD# Setup to ADCLK Rising Edge RD# Hold from ADCLK Rising Edge DQ[31:0] Output Valid from ADCLK Rising Edge DQ[31:0] Output Hold from ADCLK Rising Edge DQ[31:0] Output Float from ADCLK Rising Edge WR# Setup to ADCLK Rising Edge WR# Hold from ADCLK Rising Edge DQ[31:0] Input Setup to ADCLK Rising Edge DQ[31:0] Input Hold from ADCLK Rising Edge 8.3 1 6.7 1 16 3 15.6 7.4 1 5.4 1 ns ns ns ns ns ns ns ns ns ns ns 2 2 2 2 2 1. Refers to Pass-Thru Passive mode only. 2. Refers to Pass-Thru Active and Passive modes. AMCC Confidential and Proprietary DS1596 150 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Figure 86. Add-On Clock Timing t10 2.0 0.8 2.0 t12 VIH2 t11 TACL 2.0 0.8 2.0 0.8 t13 Figure 87. Pass-Thru Clock Relationship to PCI Clock PCI CLK t14 t15 BPCLK Figure 88. PTADR Timing PTADR# t16 t17a t17b DQ[31:0] ADDRESS AMCC Confidential and Proprietary DS1596 151 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Figure 89. Passive Mode Pass-Thru Operation ADCLK PTATN# t18 PTBURST# t20 PTNUM[1:0] t22 PTWR t24 PTBE[3:0] t26 PTRDY# t28 SELECT# t30 ADR[6:2] t32 BE[3:0] t34 RD# t36 DQ[31:0] t38 WR# t41 DQ[31:0] t43 t44 t42 Valid Out t39 t37 t40 t33 t35 t31 t29 Valid Out t27 Valid Out t25 Valid Out t23 t21 t19 AMCC Confidential and Proprietary DS1596 152 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Table 65. Add-On Timings Functional Operation Range (VCC = 5.0 V ± 5%, 0°C to 70°C, 50 pF load on outputs for MAX, 0 pF load for MIN) Symbol Parameter Min Max Units Notes t45 t46 t47 t48 t49 t50 t51 Notes: DXFER# Valid from ADCLK Rising Edge DXFER# Hold from ADCLK Rising Edge PTADR# Valid from ADCLK Rising Edge PTADR# Hold from ADCLK Rising Edge DQ[31:0] Address Valid from ADCLK Rising Edge DQ[31:0] Hold from ADCLK Rising Edge DQ[31:0] Adress Float from ADCLK Rising Edge 3 3 3 12.5 ns ns 14 ns ns 17.3 ns ns 1 1 1 16.9 ns 1. Refers to Pass-Thru Active mode only. AMCC Confidential and Proprietary DS1596 153 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Figure 90. Active Mode Pass-Thru Write Operation ADCLK PTATN# t18 PTBURST# t19 PTNUM[1:0] t22 PTWR t24 PTBE[3:0] t26 DXFER# Valid Out t23 Valid Out t25 Valid Out t27 t45 PTWAIT# t28 PTADR# t47 DQ[31:0] t49 Address t50 t51 t38 t48 Data t39 t29 t46 t40 AMCC Confidential and Proprietary DS1596 154 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Table 66. Mailbox Timings Functional Operation Range (VCC = 5.0 V ± 5%, 0°C to 70°C, 50 pF load on outputs) Symbol Parameter Min Max Units Notes t60 t61 t62 t63 t64 t65 Notes: LOAD# Setup to ADCLK Rising Edge LOAD# Hold from ADCLK Rising Edge MD[7:0] Setup to ADCLK Rising Edge MD[7:0] Hold from ADCLK Rising Edge MD[7:0] Float from LOAD# Low MD[7:0] Active from ADCLK Rising Edge 1 1 4 ns ns 2.6 ns ns 1 1 2 2,3 14.9 17 ns ns 1. Applies only when external mailbox is in input mode (MDMODE = 1). 2. Applies only when external mailbox is in input/output mode (MDMODE = 0). 3. When the S5920 is driving MD[7:0] w/ Add-On incoming mailbox byte 3, the PCI can update this output synchronously to the PCI clock. As a result, once driving, this output is asynchronous to ADCLK. Figure 91. Mailbox Data ADCLK t61 LOAD# t60 t63 MD[7:0] t62 Figure 92. Mailbox Data ADCLK t60 LOAD# t64 MD[7:0] 5920 Driving Add-On Driving t61 t62 t65 5920 Driving t63 AMCC Confidential and Proprietary DS1596 155 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Figure 93. S5920 Pinout and Pin Assignment PTBE3# PTBE2# PTBE1# DQ20 PTBE0# PTRDY# PTATN# RSVD2 PTBURST# VCC GND LOAD# PTWR PTADR# GND DQ21 PTMODE VCC ADDINT# MD7 DQ0 DQ1 DQ2 MD6 DQ3 DQ4 DQ5 DQ22 DQ6 VCC GND MD5 DQ7 BE0# DQ8 DQ23 DQ9 DQ10 DQ11 MD4 GND PTNUM1 PTNUM0 IRQ# DQ19 SYSRST# SDA SCL VCC GND VCC ADR6 DQ18 ADCLK RSVD3 RSVD4 GND FLT# RST# BPCLK GND CLK MDMODE DXFR# DQ17 AD31 AD30 AD29 RSVD5 GND VCC AD28 GND AD27 AD26 AD25 DQ16 AD24 C/BE3# IDSEL 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 120 119 118 117 116 115 114 113 112 111 110 109 108 107 106 105 104 103 102 101 100 99 98 97 96 95 94 93 92 91 90 89 88 87 86 85 84 83 82 81 S5920 160 PQFP 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 49 48 47 46 45 44 43 42 41 DQ12 DQ13 DQ14 DQ24 DQ15 SELECT# WR# MD3 RD# VCC GND MD2 ADR2 ADR3 ADR4 DQ25 ADR5 BE1# BE2# MD1 BE3# DQMODE INTA# MD0 AD0 AD1 AD2 DQ26 AD3 VCC GND GND AD4 AD5 AD6 DQ27 AD7 C/BE0# AD8 GND GND AD23 AD22 AD21 DQ31 AD20 AD19 AD18 GND GND VCC AD17 DQ30 AD16 C/BE2# FRAME# GND IRDY# TRDY# DEVSEL# GND STOP# LOCK# PERR# DQ29 SERR# PAR C/BE1# RSVD1 GND VCC AD15 VCC AD14 AD13 AD12 DQ28 AD11 AD10 AD9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 AMCC Confidential and Proprietary DS1596 156 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book S5920 – 160 PQFP PACKAGE MARKING DRAWING (TOP VIEW) Figure 94. S5920 – 160 PQFP Package Marking Drawing (Top View) NOTES (Unless Otherwise Specified): 1 2 Dot Represents PIN 1 (A01) Designator ES (Engineering Sample) designator. When present, this signifies pre-production grade material. Pre-production grade material is not guaranteed to meet the specifications in this document. 1 PCI MATCHMAKER S5920QRC YYWW ZZZZZZM JJJ TAIWAN e2 ES 2 LEGEND (in row order - including symbols): ROW #1: ROW #2: ROW #3: AMCC Logo AMCC Device Part Number Mask Protection Symbol XX: Assembly Year Code YY: Assembly Week Code ZZZZZZ: AMCC 6 Digit Lot Code M: Mask Set Revision Code JJJ: up to 3 Digit Subcontractor Lot Code ESD Symbol TAIWAN: Assembly Location e2 M ROW #4: ROW #5: ROW #6: ROW #7: ROW #8: ROW #9: RoHS Lead Free Compliant Symbol (per JEDEC: JESD97). When present, this signifies a lead free package. Engineering Sample (ES) Designator (only on Pre-Production Devices) AMCC Confidential and Proprietary DS1596 157 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Figure 95. 160 PQFP (28 x 28 x 3.37 mm) - Plastic Quad Flat Package Package Material Note: Standard Package: Pin Compiosition - 85Sn/15Pb. Green/RoHS Compliant Package: Pin Composition - 98Sn/2.0Cu AMCC Confidential and Proprietary DS1596 158 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book DOCUMENT REVISION HISTORY Revision Date Description 1.01 11/28/05 • • • • Page 2, Updated Features section Page 157, Added Figure 94, S5920 – 176 LQFP Package Marking Drawing Page 158, Added Note to Figure 95 Page 160, Updated Ordering Information AMCC Confidential and Proprietary DS1596 159 S5920 – PCI Product Revision 1.01 – November 28, 2005 Data Book Ordering Information Prefix S - Integrated Circuit S - Integrated Circuit Device 5920 5920 Package Q -160 PQFP, Standard Package QRC - 160 PQFP ,RoHS/Green Lead Free X Prefix XXXX Device XX Package Applied Micro Circuits Corporation 6290 Sequence Dr., San Diego, CA 92121 Phone: (858) 450-9333 — (800) 755-2622 — Fax: (858) 450-9885 http://www.amcc.com AMCC reserves the right to make changes to its products, its datasheets, or related documentation, without notice and warrants its products solely pursuant to its terms and conditions of sale, only to substantially comply with the latest available datasheet. Please consult AMCC’s Term and Conditions of Sale for its warranties and other terms, conditions and limitations. AMCC may discontinue any semiconductor product or service without notice, and advises its customers to obtain the latest version of relevant information to verify, before placing orders, that the information is current. AMCC does not assume any liability arising out of the application or use of any product or circuit described herein, neither does it convey any license under its patent rights nor the rights of others. AMCC reserves the right to ship devices of higher grade in place of those of lower grade. AMCC SEMICONDUCTOR PRODUCTS ARE NOT DESIGNED, INTENDED, AUTHORIZED, OR WARRANTED TO BE SUITABLE FOR USE IN LIFE-SUPPORT APPLICATIONS, DEVICES OR SYSTEMS OR OTHER CRITICAL APPLICATIONS. AMCC is a registered Trademark of Applied Micro Circuits Corporation. Copyright © 2005 Applied Micro Circuits Corporation. AMCC Confidential and Proprietary DS1596 160