User Guide
Table Of Contents
- Intel® IXP2800 Network Processor
- Copyright
- Contents
- Introduction 1
- Technical Description 2
- 2.1 Overview
- 2.2 Intel XScale® Core Microarchitecture
- 2.3 Microengines
- 2.4 DRAM
- 2.5 SRAM
- 2.6 Scratchpad Memory
- 2.7 Media and Switch Fabric Interface
- 2.8 Hash Unit
- 2.9 PCI Controller
- 2.10 Control and Status Register Access Proxy
- 2.11 Intel XScale® Core Peripherals
- 2.12 I/O Latency
- 2.13 Performance Monitor
- Intel XScale® Core 3
- 3.1 Introduction
- 3.2 Features
- 3.3 Memory Management
- 3.4 Instruction Cache
- 3.5 Branch Target Buffer (BTB)
- 3.6 Data Cache
- 3.6.1 Overviews
- 3.6.2 Data Cache and Mini-Data Cache Operation
- 3.6.3 Data Cache and Mini-Data Cache Control
- 3.6.4 Reconfiguring the Data Cache as Data RAM
- 3.6.5 Write Buffer/Fill Buffer Operation and Control
- 3.7 Configuration
- 3.8 Performance Monitoring
- 3.9 Performance Considerations
- 3.9.1 Interrupt Latency
- 3.9.2 Branch Prediction
- 3.9.3 Addressing Modes
- 3.9.4 Instruction Latencies
- 3.9.4.1 Performance Terms
- 3.9.4.2 Branch Instruction Timings
- 3.9.4.3 Data Processing Instruction Timings
- 3.9.4.4 Multiply Instruction Timings
- 3.9.4.5 Saturated Arithmetic Instructions
- 3.9.4.6 Status Register Access Instructions
- 3.9.4.7 Load/Store Instructions
- 3.9.4.8 Semaphore Instructions
- 3.9.4.9 Coprocessor Instructions
- 3.9.4.10 Miscellaneous Instruction Timing
- 3.9.4.11 Thumb Instructions
- 3.10 Test Features
- 3.11 Intel XScale® Core Gasket Unit
- 3.12 Intel XScale® Core Peripheral Interface
- 3.12.1 XPI Overview
- 3.12.2 UART Overview
- 3.12.3 UART Operation
- 3.12.4 Baud Rate Generator
- 3.12.5 General Purpose I/O (GPIO)
- 3.12.6 Timers
- 3.12.7 Slowport Unit
- Microengines 4
- DRAM 5
- SRAM Interface 6
- SHaC - Unit Expansion 7
- Media and Switch Fabric Interface 8
- 8.1 Overview
- 8.2 Receive
- 8.3 Transmit
- 8.4 RBUF and TBUF Summary
- 8.5 CSIX Flow Control Interface
- 8.6 Deskew and Training
- 8.7 CSIX Startup Sequence
- 8.8 Interface to Command and Push and Pull Buses
- 8.9 Receiver and Transmitter Interoperation with Framers and Switch Fabrics
- 8.9.1 Receiver and Transmitter Configurations
- 8.9.2 System Configurations
- 8.9.2.1 Framer, Single Network Processor Ingress and Egress, and Fabric Interface Chip
- 8.9.2.2 Framer, Dual Network Processor Ingress, Single Network Processor Egress, and Fabric Interface Chip
- 8.9.2.3 Framer, Single Network Processor Ingress and Egress, and CSIX-L1 Chips for Translation and Fabric Interface
- 8.9.2.4 CPU Complex, Network Processor, and Fabric Interface Chip
- 8.9.2.5 Framer, Single Network Processor, Co-Processor, and Fabric Interface Chip
- 8.9.3 SPI-4.2 Support
- 8.9.4 CSIX-L1 Protocol Support
- 8.9.5 Dual Protocol (SPI and CSIX-L1) Support
- 8.9.6 Transmit State Machine
- 8.9.7 Dynamic De-Skew
- 8.9.8 Summary of Receiver and Transmitter Signals
- PCI Unit 9
- 9.1 Overview
- 9.2 PCI Pin Protocol Interface Block
- 9.2.1 PCI Commands
- 9.2.2 IXP2800 Network Processor Initialization
- 9.2.3 PCI Type 0 Configuration Cycles
- 9.2.4 PCI 64-Bit Bus Extension
- 9.2.5 PCI Target Cycles
- 9.2.6 PCI Initiator Transactions
- 9.2.7 PCI Fast Back-to-Back Cycles
- 9.2.8 PCI Retry
- 9.2.9 PCI Disconnect
- 9.2.10 PCI Built-In System Test
- 9.2.11 PCI Central Functions
- 9.3 Slave Interface Block
- 9.4 Master Interface Block
- 9.5 PCI Unit Error Behavior
- 9.5.1 PCI Target Error Behavior
- 9.5.1.1 Target Access Has an Address Parity Error
- 9.5.1.2 Initiator Asserts PCI_PERR_L in Response to One of Our Data Phases
- 9.5.1.3 Discard Timer Expires on a Target Read
- 9.5.1.4 Target Access to the PCI_CSR_BAR Space Has Illegal Byte Enables
- 9.5.1.5 Target Write Access Receives Bad Parity PCI_PAR with the Data
- 9.5.1.6 SRAM Responds with a Memory Error on One or More Data Phases on a Target Read
- 9.5.1.7 DRAM Responds with a Memory Error on One or More Data Phases on a Target Read
- 9.5.2 As a PCI Initiator During a DMA Transfer
- 9.5.2.1 DMA Read from DRAM (Memory-to-PCI Transaction) Gets a Memory Error
- 9.5.2.2 DMA Read from SRAM (Descriptor Read) Gets a Memory Error
- 9.5.2.3 DMA from DRAM Transfer (Write to PCI) Receives PCI_PERR_L on PCI Bus
- 9.5.2.4 DMA To DRAM (Read from PCI) Has Bad Data Parity
- 9.5.2.5 DMA Transfer Experiences a Master Abort (Time-Out) on PCI
- 9.5.2.6 DMA Transfer Receives a Target Abort Response During a Data Phase
- 9.5.2.7 DMA Descriptor Has a 0x0 Word Count (Not an Error)
- 9.5.3 As a PCI Initiator During a Direct Access from the Intel XScale® Core or Microengine
- 9.5.3.1 Master Transfer Experiences a Master Abort (Time-Out) on PCI
- 9.5.3.2 Master Transfer Receives a Target Abort Response During a Data Phase
- 9.5.3.3 Master from the Intel XScale® Core or Microengine Transfer (Write to PCI) Receives PCI_PERR_L on PCI Bus
- 9.5.3.4 Master Read from PCI (Read from PCI) Has Bad Data Parity
- 9.5.3.5 Master Transfer Receives PCI_SERR_L from the PCI Bus
- 9.5.3.6 Intel XScale® Core Microengine Requests Direct Transfer when the PCI Bus is in Reset
- 9.5.1 PCI Target Error Behavior
- 9.6 PCI Data Byte Lane Alignment
- Clocks and Reset 10
- 10.1 Clocks
- 10.2 Synchronization Between Frequency Domains
- 10.3 Reset
- 10.4 Boot Mode
- 10.5 Initialization
- Performance Monitor Unit 11
- 11.1 Introduction
- 11.2 Interface and CSR Description
- 11.3 Performance Measurements
- 11.4 Events Monitored in Hardware
- 11.4.1 Queue Statistics Events
- 11.4.2 Count Events
- 11.4.3 Design Block Select Definitions
- 11.4.4 Null Event
- 11.4.5 Threshold Events
- 11.4.6 External Input Events
- 11.4.6.1 XPI Events Target ID(000001) / Design Block #(0100)
- 11.4.6.2 SHaC Events Target ID(000010) / Design Block #(0101)
- 11.4.6.3 IXP2800 Network Processor MSF Events Target ID(000011) / Design Block #(0110)
- 11.4.6.4 Intel XScale® Core Events Target ID(000100) / Design Block #(0111)
- 11.4.6.5 PCI Events Target ID(000101) / Design Block #(1000)
- 11.4.6.6 ME00 Events Target ID(100000) / Design Block #(1001)
- 11.4.6.7 ME01 Events Target ID(100001) / Design Block #(1001)
- 11.4.6.8 ME02 Events Target ID(100010) / Design Block #(1001)
- 11.4.6.9 ME03 Events Target ID(100011) / Design Block #(1001)
- 11.4.6.10 ME04 Events Target ID(100100) / Design Block #(1001)
- 11.4.6.11 ME05 Events Target ID(100101) / Design Block #(1001)
- 11.4.6.12 ME06 Events Target ID(100110) / Design Block #(1001)
- 11.4.6.13 ME07 Events Target ID(100111) / Design Block #(1001)
- 11.4.6.14 ME10 Events Target ID(110000) / Design Block #(1010)
- 11.4.6.15 ME11 Events Target ID(110001) / Design Block #(1010)
- 11.4.6.16 ME12 Events Target ID(110010) / Design Block #(1010)
- 11.4.6.17 ME13 Events Target ID(110011) / Design Block #(1010)
- 11.4.6.18 ME14 Events Target ID(110100) / Design Block #(1010)
- 11.4.6.19 ME15 Events Target ID(110101) / Design Block #(1010)
- 11.4.6.20 ME16 Events Target ID(100110) / Design Block #(1010)
- 11.4.6.21 ME17 Events Target ID(110111) / Design Block #(1010)
- 11.4.6.22 SRAM DP1 Events Target ID(001001) / Design Block #(0010)
- 11.4.6.23 SRAM DP0 Events Target ID(001010) / Design Block #(0010)
- 11.4.6.24 SRAM CH3 Events Target ID(001011) / Design Block #(0010)
- 11.4.6.25 SRAM CH2 Events Target ID(001100) / Design Block #(0010)
- 11.4.6.26 SRAM CH1 Events Target ID(001101) / Design Block #(0010)
- 11.4.6.27 SRAM CH0 Events Target ID(001110) / Design Block #(0010)
- 11.4.6.28 DRAM DPLA Events Target ID(010010) / Design Block #(0011)
- 11.4.6.29 DRAM DPSA Events Target ID(010011) / Design Block #(0011)
- 11.4.6.30 IXP2800 Network Processor DRAM CH2 Events Target ID(010100) / Design Block #(0011)
- 11.4.6.31 IXP2800 Network Processor DRAM CH1 Events Target ID(010101) / Design Block #(0011)
- 11.4.6.32 IXP2800 Network Processor DRAM CH0 Events Target ID(010110) / Design Block #(0011)
112 Hardware Reference Manual
Intel
®
IXP2800 Network Processor
Intel XScale
®
Core
3.9.1 Interrupt Latency
Minimum Interrupt Latency is defined as the minimum number of cycles from the assertion of any
interrupt signal (IRQ or FIQ) to the execution of the instruction at the vector for that interrupt. The
point at which the assertion begins is TBD. This number assumes best case conditions exist when
the interrupt is asserted, e.g., the system isn’t waiting on the completion of some other operation.
A useful number to work with is the Maximum Interrupt Latency. This is typically a complex
calculation that depends on what else is going on in the system at the time the interrupt is asserted.
Some examples that can adversely affect interrupt latency are:
• The instruction currently executing could be a 16-register LDM.
• The processor could fault just when the interrupt arrives.
• The processor could be waiting for data from a load, doing a page table walk, etc.
• There are high core-to-system (bus) clock ratios.
Maximum Interrupt Latency can be reduced by:
• Ensuring that the interrupt vector and interrupt service routine are resident in the instruction
cache. This can be accomplished by locking them down into the cache.
• Removing or reducing the occurrences of hardware page table walks. This also can be
accomplished by locking down the application’s page table entries into the TLBs, along with
the page table entry for the interrupt service routine.
3.9.2 Branch Prediction
The Intel XScale
®
core implements dynamic branch prediction for the ARM* instructions B and
BL and for the Thumb instruction B. Any instruction that specifies the PC as the destination is
predicted as not taken. For example, an LDR or a MOV that loads or moves directly to the PC will
be predicted not taken and incur a branch latency penalty.
These instructions -- ARM B, ARM BL and Thumb B -- enter into the branch target buffer when
they are “taken” for the first time. (A “taken” branch refers to when they are evaluated to be true.)
Once in the branch target buffer, the Intel XScale
®
core dynamically predicts the outcome of these
instructions based on previous outcomes. Table 26 shows the branch latency penalty when these
instructions are correctly predicted and when they are not. A penalty of 0 for correct prediction
means that the Intel XScale
®
core can execute the next instruction in the program flow in the cycle
following the branch.
Table 26. Branch Latency Penalty
Core Clock Cycles
Description
ARM* Thumb
+0 + 0
Predicted Correctly. The instruction is in the branch target cache and is
correctly predicted.
+4 + 5
Mispredicted. There are three occurrences of branch misprediction, all of which
incur a 4-cycle branch delay penalty.
1. The instruction is in the branch target buffer and is predicted not-taken, but is
actually taken.
2. The instruction is not in the branch target buffer and is a taken branch.
3. The instruction is in the branch target buffer and is predicted taken, but is
actually not-taken