Site affiché en Français Voir le site en English (USA) Voir le site en English (GB)
Vous êtes ici: ac6 > ac6-formation > Marvell bridges > MARVELL MV64660 implementation
Télécharger le catalogue
Télécharger le catalogue
Télécharger la page
Télécharger la page
Ecrivez nous
Ecrivez nous
Version imprimable
Version imprimable
 

MV3 MARVELL MV64660 implementation

This course covers Marvell Discovery VI devices


formateur
Objectives
bullet_jaune_1 The course describes the MV64660 internal data paths.
bullet_jaune_1 The course explains how the host PowerPC and a CPU connected to PCI-X can synchronize to each other through the message unit.
bullet_jaune_1 Operation of the PCI Express interface is detailed in Root Complex mode as well as in Endpoint mode.
bullet_jaune_1 A long introduction to DDR SDRAM is done prior to describe the DDR SDRAM controller operation.
bullet_jaune_1 The course focuses on the hardware implementation of the DDR SDRAM.
bullet_jaune_1 The training explains how to implement chained DMA transfers, by using either IDMA channels or XOR engines.
bullet_jaune_1 The course highlights the possible optimizations that can be implemented to boost the performance of the Ethernet controller.
A more detailed course description is available on request at info@ac6-formation.com
Pre-requisites
bullet_jaune_2 Knowledge of PowerPC 60X / MPX bus. See our courses on Freescale and IBM Microelectronics PowerPCs.
Related courses
bullet_jaune_2 Ethernet and switching, reference N1
bullet_jaune_2 PCI express, reference IC4
bullet_jaune_2 USB Full Speed High Speed, reference IP2
bullet_jaune_2 Serial-ATA, reference IS3


Plan
OVERVIEW
bullet_jaune_2 6-bus architecture, organization of a board based on MV64660
bullet_jaune_2 Frequency domains, fast path between CPU and SRAM / SDRAM
bullet_jaune_2 Data integrity checking
bullet_jaune_2 Internal crossbar
bullet_jaune_2 Master de-mux programming, address decode windows
bullet_jaune_2 Slave mux programming, pizza arbiters operation
CPU INTERFACE
bullet_jaune_2 CPU address space decoding
bullet_jaune_2 CPU-to-PCI address remapping
bullet_jaune_2 Protection windows
bullet_jaune_2 Arbitration, multi-processor operation
bullet_jaune_2 CPU slave operation
bullet_jaune_2 Cache coherency
bullet_jaune_2 Deadlock avoidance
bullet_jaune_2 Transaction ordering
bullet_jaune_2 Hardware implementation, clocking, low power modes
DDR2 INTERFACE
bullet_jaune_2 Introduction to DDR SDRAM from Jedec specification
bullet_jaune_2 DDR2 on-die terminations
bullet_jaune_2 Clocking
bullet_jaune_2 Initialization sequence
bullet_jaune_2 Data synchronization : DQS signals, programmable DLL
bullet_jaune_2 DDR2 SDRAM controller, functional description
bullet_jaune_2 Page management
bullet_jaune_2 Read and write transactions
bullet_jaune_2 ECC and read-modify-write transactions
bullet_jaune_2 Hardware implementation, ODT management (internal and external)
bullet_jaune_2 Low power modes
DEVICE CONTROLLER
bullet_jaune_2 Transaction queue, read and write data buffers
bullet_jaune_2 Address and data multiplexing
bullet_jaune_2 Timing parameters
bullet_jaune_2 External acknowledgement
bullet_jaune_2 Pack / unpack and burst support
bullet_jaune_2 NAND flash support, boot from NAND flash
PCI INTERFACE
bullet_jaune_2 PCI bus arbitration
bullet_jaune_2 Master operation in PCI and PCI-X mode
bullet_jaune_2 Target operation in PCI and PCI-X mode
bullet_jaune_2 PCI-to-PCI configuration transactions
bullet_jaune_2 Address decoding
PCI-EXPRESS x4 AND x1 INTERFACES
bullet_jaune_2 Integrated low power SERDES PHY
bullet_jaune_2 x1, x4 link
bullet_jaune_2 Operating as either Root Complex or Endpoint
bullet_jaune_2 Link initialization
bullet_jaune_2 Arbitration and ordering
bullet_jaune_2 Peer-to-peer traffic
bullet_jaune_2 Messaging unit, synchronization between CPUs through PCI/PCI-Express
GENERAL PURPOSE INPUT/ OUTPUT PINS
bullet_jaune_2 GPIO port, functional description
bullet_jaune_2 Interrupt request inputs
bullet_jaune_2 Multi Purpose Pin multiplexing
INTERRUPT CONTROLLERS AND TIMERS
bullet_jaune_2 Watchdog timer
bullet_jaune_2 Interrupt controller functional description
bullet_jaune_2 Interrupt steering logic to 4 possible output pins
bullet_jaune_2 Priority mechanism
TWSI CONTROLLER AND RESET
bullet_jaune_2 Master and slave operation, 7- or 10-bit addressing
bullet_jaune_2 Determining the current state of the controller by reading the status register
bullet_jaune_2 Master write sequence, master read sequence
bullet_jaune_2 Slave write sequence, slave read sequence
bullet_jaune_2 Reset pins and configuration
bullet_jaune_2 Utilization of the boot sequencer
bullet_jaune_2 Requirement for an external Central Resource CPLD
IDMA CHANNELS
bullet_jaune_2 IDMA address decoding
bullet_jaune_2 Target unit and attributes programming
bullet_jaune_2 Functional description, external control
bullet_jaune_2 Normal mode vs chained mode
bullet_jaune_2 Transfer descriptors, descriptor ownership
bullet_jaune_2 Channel activation
bullet_jaune_2 DMA interrupts
XOR ENGINES
bullet_jaune_2 State machine : Active, Inactive and Paused states
bullet_jaune_2 XOR, CRC and DMA operation modes, format of transfer descriptors
bullet_jaune_2 XOR operation mode
bullet_jaune_2 CRC32 operation mode
bullet_jaune_2 DMA operation mode
bullet_jaune_2 Memory Initialization operation mode
bullet_jaune_2 ECC error cleanup operation mode
bullet_jaune_2 Arbitration between XOR engine0 and XOR engine1
bullet_jaune_2 Address override capability
bullet_jaune_2 XOR Engines interrupts
16550 COMPATIBLE UARTs
bullet_jaune_2 FIFO mode
bullet_jaune_2 Flow control
bullet_jaune_2 Transmit sequence
bullet_jaune_2 Receive sequence
USB2.0 PORTS
bullet_jaune_2 Address decoding
bullet_jaune_2 Integrated PHY
bullet_jaune_2 USB host operation, EHCI specification support
bullet_jaune_2 USB device operation, Endpoint configuration
bullet_jaune_2 Dedicated DMA for data movement between memory and port
SATA-II INTERFACE
bullet_jaune_2 Integrated PHY, 3.0 or 1.5 Gbps bit rate
bullet_jaune_2 EDMA request and response queues
bullet_jaune_2 Studying the sequence that the software must implement to perform a PIO transfer
bullet_jaune_2 Studying the sequence that the software must implement to perform a DMA transfer
bullet_jaune_2 Queued DMA
bullet_jaune_2 Interrupt coalescing
bullet_jaune_2 Port multiplier support
GIGABIT ETHERNET CONTROLLERS
bullet_jaune_2 SGMII support
bullet_jaune_2 Dedicated DMA
bullet_jaune_2 Related interrupts
bullet_jaune_2 Transmit weighted round-robin arbitration
bullet_jaune_2 Backpressure mode
bullet_jaune_2 Transmit and receive sequences
bullet_jaune_2 Management interface
bullet_jaune_2 MIB
CRYPTOGRAPHIC ENGINE AND SECURITY ACCELERATORS
bullet_jaune_2 Cryptographic engine functional description
bullet_jaune_2 Involved units : CPU, dedicated SRAM and security accelerator
bullet_jaune_2 Authentication
bullet_jaune_2 Encryption and decryption, supported algorithms
bullet_jaune_2 TDMA controller, attaching TDMA to security accelerator
bullet_jaune_2 Multi-packet chained mode
bullet_jaune_2 DES encryption / decryption sequence, pipelining