Happy 10th Anniversary to Open-Silicon!

Open-Silicon is celebrating another birthday! It has been 10 years since Open-Silicon began serving the semiconductor community here in the Silicon Valley and around the globe.

Differentiate Using Customizable, Configurable IP

Today’s products require more and more features and functionality to stand apart. As a silicon proven solution, Open-Silicon IP can be customized to enable further customer product differentiation. From the Interlaken high-performance chip-to-chip interconnect through SoC platform IP supporting different processor cores and peripherals, error-correction codes, compression engines, security and I/O controller IP solutions, Open-Silicon provides reliable high-quality IP solutions with a rapid path to silicon success.

IP listed here is available as stand-alone third-party IP, or as a part of the customizable system and physical design solutions offered by Open-Silicon.


Interlaken Controller IP

Interlaken

Open-Silicon’s Interlaken IP is a scalable, high-speed chip-to-chip interface protocol that builds on the channelization and per-channel flow control features of SPI4.2, while reducing the number of chip I/O pins by using high-speed SerDes technology, similar to XAUI. The Interlaken IP supports the following
Interlaken Alliance specifications:

  • Interlaken Protocol Definition, v1.2
  • Interlaken Look-Aside Protocol Definition, v1.1
  • Interlaken Retransmit Extension, v1.1
  • Interlaken Interop Recommendations, v1.6

Designed and tested to be easily synthesizable into many ASIC technologies, Open-Silicon’s Interlaken IP Core was uniquely built to work with off-the-shelf SerDes from leading technology vendors. Using vendor specific, proven, SerDes allows Open-Silicon customers to quickly integrate the Interlaken IP Core into the customer’s technology of choice.

Open-Silicon’s sixth-generation Interlaken IP core doubles user data bandwidth of the previous generation to 600Gbps by enhancing the SerDes and user interface support. This version of the IP core improves system reliability with the addition of Interlaken Retransmit Extension support. Building upon Open-Silicon’s robust and flexible architecture, the IP also adds support for multiple aggregate bandwidth interfaces within a single IP core instance, allowing for a more efficient implementation. 

Key Features:

  • Fully-programmable SerDes lane mapping
  • 4-channel Interlaken Look Aside protocol support
  • 25Gbps SerDes support
  • 600Gbps high-bandwidth performance
  • Interlaken Retransmit Extension support

In addition to the key features highlighted with the latest release, the Open-Silicon Interlaken IP also provides the following feature set as part of the standard IP functionality:

  • Support for 256 logical channels, plus 8 bit channel extension for up to 64K channels
  • Independent SerDes lane enable/disable
  • Support for SerDes speeds from 3.125Gbps to 25Gbps
  • Configurable number of lanes from 1 to 48
  • Three options for User Interface: 128, 2×128, 256, 2×256 and 4×256-bit
  • Programmable BURSTMAX from 64 bytes – 512 bytes
  • Programmable BURSTMIN from 32 bytes – 256 bytes
  • In-band and Out-of-Band flow control
  • Programmable calendar
  • Built-in error detection and interrupt structures
  • Configurable error injection mechanisms for testability
  • Full-packet mode and segment mode
  • SerDes support for 8, 10, 16, 20 or 32-bit SerDes
  • Maintenance interface for control and configuration

Interlaken White Paper

Download link coming soon

Interlaken Product Brochure

Download link coming soon

SoC Platform IP Solutions

The Open-Silicon SoC platforms are highly-configurable system on chip (SoC) solutions. The SoC platforms are pre-verified, integrated and customized platforms which include a rich suite of standard infrastructure IP. Other 3rd party IP blocks can easily be integrated into the SoC platforms. The platforms are flexible and in the case of the SoC infrastructure subsystem, it is processor independent and easily integrated with processor such as ARM, ARC, MIPS, Tensilica and other processor or DSP cores. The solutions include a robust verification platform that allows for HW/SW co-verification.

MCU SoC Platform Solutions

The MCU SoC platform solution consists a highly-configurable SoC platform solution.  The SoC platform solution includes  8051 SoC platform and 80251 SoC Platform which is 8051 CPU and 80251 CPU integrated SoC platform. The solution includes a robust verification platform that allows for HW/SW co-verification.

Configurable SoC Platform with 8051

Description

The SOCC-1110 is an 8051 based configurable SoC platform. The platform contains an 8051 CPU core, and all the peripheral functions required for a basic SoC.

The SOCC-1110  contains a single cycle 8051 CPU core plus standard set of SoC peripherals, including a system timer, a watchdog timer, an interrupt controller, programmable I/O, I2C Host, SPI master, real-time clock, alphanumeric LCD interface, 16550 UART and serial port. A memory controller supporting SRAM, Serial Flash, and Serial EERPROM also be optionally instantiated. The user also has the option to select from a library of cores and subsystems that have been pre-integrated and verified to work in this platform, including a DMA controller, 16450 UART, IDE Host, SATA Host and SATA Device interface. Also, 3rd party IP cores like JPEG decoders, TCP/IP hardware accelerators and Ethernet MACs can be easily integrated into this platform.

Many applications can be supported by this configurable platform including Home Area Networks (HAN), Smart Grid, Zigbee, wireless sensor networks, USB Flash Drives, USB IDE/SATA Drives, Web servers, Smart Card, Photo Frames etc.

Features

  • 8051 single cycle CPU core
  • Standard peripheral functions
    • Programmable I/O
    • 16450/16550 compatible UART functions
    • Interrupt controller
    • General purpose timer
    • Watchdog timer
    • Power management
    • Real-time clock
    • I2C master
    • SPI master
    • Alphanumeric LCD controller
  • Optional component available
    • DMA controller
    • Memory controller for SRAM/Flash/EEPROM
    • MMC
  • Other 3rd party IP blocks can easily be integrated

Applications

The SOCC-1110  SoC platform may form the basis for a complete chip or it may also serve as a subsystem in a larger chip. Peripheral blocks may be added as needed, with the CPU and memory subsystems unaffected.

The example illustrated above features the SOCC-1110 core as a USB to SATA bridge.  A USB core, SATA host controller, and DMA engine are added to the baseline platform to make it an end market driven platform.

Deliverables

  • Verilog RTL code
  • Synthesis/STA scripts
  • Comprehensive documentation
    • Datasheet
    • User Manual
    • Memory-mapped register manual
    • DV plan
    • Release notes
  • Verification deliverables
    • Assembly language test suites for CPU verification
    • Assembly language test suites for peripheral/platform verification
    • C test suites to verify peripherals/platforms
    • Customized scripts to compile programs and load memories as required
    • Verilog BFM/test suites for platform verification
    • Same test suites run at module-level or platform-level
    • Includes simulation models for all peripherals
    • Supports C, Verilog, assembly language based verification
  • Supports HW/SW co-verification
    • Compiler, simulator and MON51 support in Keil™
  • Support for FPGA-based emulation
  • Benchmark tests
    • Sieve
    • Dhrystone v1.1, v2.1
    • Whetstone
  • Operating systems ported to platform
    • TinyOS
    • Contiki OS

Value Proposition

  • Accelerate design-cycle by reducing development time by up to 40%
    • Allows customers to focus on developing their value-add blocks
  • Multiple peripherals available in addition to base platform to target following applications:
    • Home Area Networks (HAN)
    • Smart grid
    • Zigbee
    • Wireless sensor networks
    • USB Flash drives
    • USB IDE/SATA drives
    • Web servers
    • Smart card
    • Photo frames, etc.
  • Easy to integrate 3rd party IP
  • Robust verification platform/deliverables
  • Silicon proven IP
  • Additional Customization options
    • New IP development and integration
    • Application software and firmware development

Download

Download link coming soon

Configurable SoC Platform with 80251

Description

The SOCC-2110 is an 80251 based configurable SoC platform. The platform contains an 80251 CPU core, and all the peripheral functions required for a basic SoC.

The SOCC-2110 contains a single cycle 80251 CPU core plus standard set of SoC peripherals, including a system timer, a watchdog timer, an interrupt controller, programmable I/O, I2C Host, SPI master, real-time clock, alphanumeric LCD interface, 16550 UART and serial port. A memory controller supporting SRAM, Serial Flash, and Serial EERPROM also be optionally instantiated. The user also has the option to select from a library of cores and subsystems that have been pre-integrated and verified to work in this platform, including a DMA controller, 16450 UART, IDE Host, SATA Host and SATA Device interface. Also, 3rd party IP cores like JPEG decoders, TCP/IP hardware accelerators and Ethernet MACs can be easily integrated into this platform.

Many applications can be supported by this configurable platform including Home Area Networks (HAN), Smart Grid, Zigbee, wireless sensor networks, USB Flash Drives, USB IDE/SATA Drives, Web servers, Smart Card, Photo Frames etc.

Features

  • 80251 single cycle CPU core
  • Standard peripheral functions
    • Programmable I/O
    • Full duplex serial port
    • 16550 compatible UART
    • Interrupt controller
    • General-purpose timer
    • Watchdog timer
    • Power management
    • Real-time clock
    • I2C master
    • SPI master
    • Alphanumeric LCD interface
  • Optional component available
    • DMA controller
    • Memory controller for SRAM/Flash/EEPROM
    • MMC
  • Other 3rd party IP blocks can easily be integrated

Applications

The SOCC-2110  SoC platform may form the basis for a complete chip or it may also serve as a subsystem in a larger chip. Peripheral blocks may be added as needed, with the CPU and memory subsystems unaffected.

The example illustrated above features the SOCC-2110 core as a USB to SATA bridge.  A USB core, SATA host controller, and DMA engine are added to the baseline platform to make it an end market driven platform.

Deliverables

  • Verilog RTL code
  • Synthesis/STA scripts
  • Comprehensive documentation
    • Datasheet
    • User manual
    • Memory-mapped register manual
    • DV plan
    • Release notes
  • Verification deliverables
    • Assembly language test suites for CPU verification
    • Assembly language test suites for peripheral/platform verification
    • C test suites to verify peripherals/platforms
    • Customized scripts to compile programs and load memories as required
    • Verilog BFM/test suites for platform verification
    • Same test suites run at module-level or platform-level
    • Includes simulation models for all peripherals
    • Testbench
    • Supports C, Verilog, assembly language based verification
  • Supports HW/SW co-verification
    • Compiler, simulator and MON51 support in KeilTM
  • Support for FPGA-based emulation
  • Benchmark tests
    • Sieve
    • Dhrystone v1.1, v2.1
    • Whetstone
  • Operating systems ported to platform
    • TinyOS
    • Contiki OS

Value Proposition

  • Accelerate design-cycle by reducing development time by up to 40%
    • Allows customers to focus on developing their value-add blocks
  • Multiple peripherals available in addition to base platform to target following applications:
    • Home Area Networks (HAN)
    • Smart grid
    • Zigbee
    • Wireless sensor networks
    • USB Flash drives
    • USB IDE/SATA drives
    • Web servers
    • Smart card
    • Photo frames, etc.
  • Easy to integrate 3rd party IP
  • Robust verification platform/deliverables
  • Silicon proven IP
  • Additional customization options
    • New IP development and integration
    • Application software and firmware development

Download

Download link coming soon

MCU Processor Cores

Open-Silicon IP portfolio includes 8051 and 80251 CPU cores. The CPUs can be provided as stand-alone IP blocks or as a part of the integrated configurable SoC platform.  The deliverables include a rich suite of assembly language, C and Verilog tests.

8051 CPU Core

Description

The 8051 CPU IP core is an instruction set compatible implementation of the MCS-51 family. The 8051 CPU IP core  executes the instruction in a single clock cycle instead of the original 12 cycles per instruction execution time. The program memory has been increased to 256Kbytes, and RAM has been increased to 256 bytes. The I/O ports of the IP core have been simplified by removing unnecessary muxing on the ports to make it fully compatible with today’s system on chip (SoC) design practices.

The 8051 CPU has been designed for integration into SoC based designs. Numerous standard peripherals can be integrated to the 8051 CPU IP core by using the AMBA interface.

Features

  • Single clock cycle instruction execution
  • Up to 8Kbytes of on-chip program memory
  • 256 bytes of on-chip data RAM
  • Up to 64Kbytes of external program memory
  • Up to 248Kbytes of external data memory
  • Dual data pointers
  • Wait state support for slow external peripherals
  • Standard 2 16-bit timers/counters
  • 6 source/5 vector interrupt structure with two priority levels.
  • No multiplexed I/O ports
  • 31 general purpose I/O ports
  • Full duplex serial port
  • Optional components are available as part of 8051 platform
  • Available as library component for Open-Silicon SoC platform
  • Compliant to AMBA 2.0 specifications

Applications

The 8051 CPU IP core platform can be integrated with numerous silicon proven peripherals developed by Open-Silicon or other third party vendors to form a complete chip level solution. The 8051 CPU IP  core can be used in various application areas

  • Home Area Networks (HAN)
  • Smart grid
  • Wireless sensor network
  • ZigBee
  • USB based FLASH drives
  • USB based SATA drives

Deliverables

The IP core deliverable includes RTL code, synthesis scripts, test benches, and verification test suites.

Download

Download link coming soon

80251 CPU Core

Description

The 80251 CPU IP core is an instruction set compatible implementation of the MCS-251 family. The 80251 core executes the instruction in a single clock cycle, and is on average about 3.19 times faster than the original implementation. The performance improvements are due to architectural enhancements done to the CPU core. The I/O ports of the IP core have been simplified by removing unnecessary muxing on the ports to make it fully compatible with today’s system on chip (SoC) design practices.

The 80251 CPU has been designed for integration into SoC based designs. Numerous standard peripherals can be integrated to the 80251 core by using the AMBA interface.

Features

  • Single clock cycle instruction execution
  • 16Kbytes of on-chip program memory
  • 1024 bytes of on-chip data RAM
  • Up to 64Kbytes of external program memory
  • Up to 16Mbytes of external data memory reserve for ROM/RAM
  • Enhanced architecture for performance improvements
  • Wait state support for slow external peripherals
  • Standard 2 16-bit timers/counters
  • 6 source/5 vector interrupt structure with two priority levels
  • No multiplexed I/O ports
  • 31 general purpose I/O ports
  • Full duplex serial port
  • Optional components are available as part of 80251 platform
  • Available as library component for Open-Silicon SoC platform
  • Compliant to AMBA 2.0 specifications

Applications

The 80251 IP core platform can be integrated with numerous silicon proven peripherals developed by Open-Silicon or other third party vendors to form a complete chip level solution. The 80251 IP core can be used in various application areas:

  • Home Area Networks (HAN)
  • Smart grid
  • Wireless Sensor Network
  • ZigBee
  • USB based FLASH/SATA drives
  • Automobile
  • Home automation and security
  • High-speed modems
  • Voice applications
  • Energy measurement and saving
  • Remote less key entry
  • Toys
  • Low-power adhoc mobile networks.
  • Advanced self-adaptable sensors
  • Smart card readers
  • Medical patient monitoring

Deliverables

The IP core deliverable includes RTL code, synthesis scripts, test benches, and verification test suites.

Download

Download link coming soon

SoC Infrastructure Subsystem IP

The Open-Silicon IP portfolio consists of a rich-suite of peripherals such as timers, real-time clock, programmable I/O, UART, etc. The peripherals are compliant to AMBA AHB v2.0 bus and can be provided as stand-alone IP blocks or can be provided as part of the integrated Open-Silicon SoC Infrastructure Subsytem Solution.

SoC Infrastructure Subsystem

Description

The SOCP-1010 is a completely integrated subsystem of infrastructure IP that is pre-verified and integrated with all basic peripheral functions to enable any system on chip (SoC) design. Additionally Open-Silicon can further customize or configure this subsystem of IP based on customer requirements. This infrastructure subsystem IP is inherently processor independent and currently supports MIPS, ARC, ARM, Tensilica embedded processor cores. Other processor cores like PowerPC and DSP can easily be integrated.

The SOCP-1010 contains a standard set of SoC peripherals, including a system timer, a watchdog timer, an interrupt controller, programmable I/O, I2C Host, SPI master, 16550 and 16450 UARTs, real-time clock and an optional high-bandwidth memory controller supporting Flash, SRAM, SDRAM or DDR2/3 can also be optionally instantiated. The user also has the option to select from a library of cores and subsystems that have been pre-integrated and verified to work in this platform, including an IDE Host, SATA Host and SATA Device interface. Also, 3rd party IP cores like decoders, Gigabit Ethernet MACs, and 802.11a, 802.11b or 802.16 wireless MACs or others can be easily integrated into this platform.

Many applications can be supported by this configurable platform including MIMO mobility devices, Access Point controllers for Network Security Processors, Home Gateways, DVD players, Network routers, Set-top boxes, Storage Area Network controllers, Network Attached Storage devices, etc.

Features

  • Processor independent architecture
    • Supports MIPS, ARM, ARC, Tensilica embedded processor core
  • Standard peripheral functions
    • Programmable I/O
    • 16450/16550 compatible UART functions
    • Interrupt Controller
    • General Purpose Timer
    • Watchdog Timer
    • Power Management
    • Real Time Clock
    • I2C Master
    • SPI Master
    • Alphanumeric LCD controller
  • Optional components available
    • High-performance memory controllers for DDR2/3, SRAM/Flash
    • SATA Controller
  • Highly configurable to support a number of different application platforms like MIMO mobility devices, Storage Area Network controller, Home Gateways
  • Other 3rd party IP/blocks can easily be integrated

Applications

The SOCP-1010 SoC Infrastructure Subsystem may form the basis for a complete chip. It may also serve as a subsystem in a larger chip. Peripheral blocks may be added as needed, with the CPU and memory subsystems unaffected.

The example illustrated above features the SOCP-1010 subsystem IP core as an enhanced system for use in a hand-held  organizer. A processor, LCD interface, PC Card interface and IrDA controller are added to the basic system.

Value Proposition

  • Accelerate design-cycle by reducing development time
    • Allows customers to focus on developing their value-add blocks
  • Easy to integrate 3rd party IP
  • Silicon proven IP
  • Robust verification platform/deliverables
  • Application software and firmware development
    • Additional customization options
    • New IP development and integration

Deliverables

  • Verilog RTL code
  • Synthesis/STA scripts
  • Comprehensive documentation
    • Datasheet
    • User manual
    • Memory-mapped register manual
    • DV plan
    • Release notes
  • Verification deliverables
    • Processor independent C test suites for platform verification
    • Customized scripts to compile programs and load memories as required
    • Verilog BFM/test suites for platform verification
    • Same test suites run at module-level or platform-level
    • Simulation models and test benches included for all peripherals
    • Supports C, Verilog based verification
  • Supports HW/SW co-verification
  • Support for FPGA-based emulation

Download

Download link coming soon

Memory Controllers

Open-Silicon memory controllers include a Flash/SDRAM memory controller and DDR2 memory controller. The Flash/SDRAM shared memory controller provides a flexible interface to external single data rate synchronous DRAM and asynchronous memories such as Flash and SRAM.

DDR/DDR2 DRAM Shared Memory Processor

Description

The DDRC-1010 Shared Memory Processor provides a flexible interface to external Double Data Rate Synchronous DRAM (DDR DRAM) memories.  It includes a configurable arbitration unit and an available arbitration watchdog timer.  The DDRC-1010 has a single control port for configuration; control and status register access, and can be configured to support multiple concurrent channels. A dedicated CPU channel is provided to minimize CPU latency. Data width can be 32-bit or 64-bit.

The DDRC-1010 can be configured for up to 4 external memory banks.  Each external memory bank can access up to 1 GB of memory, and is memory-mapped with configurable address decodes.  Each bank has independent programmable timing controls, and can be configured to support either 32-bit or 64-bit data widths.

The DDRC-1010 supports 400 MHz IP core operation (800 Mbps), and can interface to AMBA buses.

Features

  •   Up to 400 MHz operation
  •   Compliant to JEDEC JESD79-2F standard
  •   Supports up to 4 external banks with up to 1GB per bank
  •   Integrated data buffers for each DMA channel
  •   Independently programmable timings for all memory banks
  •   Open bank management for 4 open pages per bank
  •   Burst controller handles burst lengths from 1 to 128
  •   SDRAM self-refresh and auto refresh, and automated SDRAM initialization
  •   Multiport concurrent channels for non-CPU masters
  •   Dedicated CPU channel to minimize CPU latency
  •   Compliant to AMBA 2.0 specifications

Applications

The DDR DRAM Shared Memory Processor provides memory access to a number of off-chip memories for multiple requestors.  All accesses are arbitrated and decoded from memory-mapped addresses by the Shared Memory Processor.

The example above shows a system with a Von Neumann architecture CPU (host processor) and Harvard architecture DSP, and a number of DMA peripheral interface blocks.  This is a standard configuration for systems with shared memory.  In the illustration, each memory requestor has an AXI interface; the Shared Memory Processor arbitrates internally between all requestors.  A single APB interface allows the CPU to configure and control the memory processor and the external memories.  An optional interrupt is available to alert the processor of a watchdog timeout, if desired. The memory processor controls access to multiple banks of DDR/DDR2 DRAM, providing fast access to large amounts of buffer memory.  A second Flash/SDRAM Shared Memory Processor provides access to Flash for processor boot-up; it is isolated so that its low performance does not impact the DDR throughput, and because of the electrical differences between the interfaces.

Deliverables

The IP core deliverable includes RTL code, synthesis scripts, test benches, and verification test suites.

Download

Download link coming soon

SDRAM/Flash Multi-port Memory Controller

Description

The FSDM-1010 Multi-port Memory Controller provides a flexible interface to external Single Data Rate Synchronous DRAM (SDRAM) and asynchronous memories such as Flash and SRAM.  It can arbitrate between up to 8 DMA requesters, with configurable arbitration and an available arbitration watchdog timer.  The FSDM-1010 can be configured for up to 8 external memory banks.  Each external memory bank can access up to 512 MB of memory, and is memory-mapped with configurable address decodes.  Each bank has independent programmable timing controls, and can be programmed to support either 16-bit or 32-bit data widths.

The SDRAM controller incorporates automated SDRAM initialization, auto-refresh, self-refresh, and open bank management.  The SRAM/Flash controller has pin- and register-configurable bus width controls, and includes facilities for a wait/ready signal for interfacing to peripheral devices.

The FSDM-1010 supports up to 166 MHz operation with 0.18-micron technology, and is available with interfaces to ARM’s AMBA buses.

This product can be integrated into the PF-1000 Open-Silicon SoC Platform as a library component.

Features

  • Up to 166 MHz operation
  • Supports up to 8 external banks with up to  512 MB per bank
  • Memory type configurable per bank as FLASH, SRAM, or SDRAM
  • Independently programmable timings for all memory types and banks
  • Open bank management for 4 open pages per bank
  • Burst controller handles burst lengths from 1 to 128
  • SDRAM self-refresh and auto-refresh
  • Automated SDRAM initialization
  • Configurable for up to 8 buffered or bufferless DMA ports
  • Configurable round robin or fixed arbitration
  • Programmable SDRAM CAS latency of 2 or 3
  • Compliant to AMBA 2.0 specifications

Application

The Flash/SDRAM Multi-port Memory Controller provides access for multiple requestors to a variety of off-chip memories.  All accesses are arbitrated and decoded from memory-mapped addresses by the shared memory processor.

The example above shows a system with a Von Neumann architecture CPU (host processor) and Harvard architecture DSP, and a number of DMA peripheral interface blocks.  This is a standard configuration for systems with shared memory.  In the illustration, each memory requestor has an AHB interface; the Multi-port Memory Controller arbitrates internally between all requestors.  A single APB interface allows the CPU to configure and control the memory processor and the external memories.  An optional interrupt is available to alert the processor of a watchdog timeout, if desired.  The memory processor also controls access to multiple banks and multiple types of memories, including Flash, SRAM and SDRAM. The physical connection to these memories may vary with the system’s requirement, but may include, for example, a single bank of 16-bit boot Flash, a single SDRAM-like device, and 4 banks of 32-bit SDRAM modules.

Deliverables

The IP core deliverable includes RTL code, synthesis scripts, test benches, and verification test suites.

Download

Download link coming soon

LCD Controller

Description

The LCDC-1010 is designed to communicate and control 4 and 8-bit character LCD devices. Primarily, LCDC-1010 is designed to support Hantronix character LCD devices but its programmable registers make it compatible with various LCD formats through different combinations. The LCDC-1010 is capable of handling two back-to-back data write transfers to the device but it can handle only single data read transfer from device.

The LCDC-1010 is designed to connect to the AMBA bus, and can  interface to most system buses. It can be integrated into the SOCP-1010 Open-Silicon SoC Platform as a library component.

Features

  • Supports 4-bit and 8-bit character LCD  interface
  • Programmable enable polarity
  • Programmable chip select
  • Programmable register strobe polarity
  • Programmable read/write polarity
  • Programmable setup time between register strobe and enable
  • Programmable enable pulse width
  • Programmable hold time
  • Interrupts/status information
    • Status bit to indicate completion of LCD transfer and busy state
    • Interrupt bits for both control register and display data - overflow, underflow
    • Masked and unmasked interrupt bits
  • Compliant to AMBA 2.0 specifications

Applications

The LCDC-1010 can be used in embedded systems to display information on LCD device in the form of text and small icons. The character LCD display are specifically designed for displaying basic text which makes it less expensive, less complex and easier to use. Character LCD displays come in various sizes, color, and background, and through the control interface of this IP core the LCD display can be reprogrammed to switch between various font sizes and colors.

Deliverables

The IP core deliverable includes RTL code, synthesis scripts, test benches, and verification test suites.

Download

Download link coming soon

ISO7816 Smart Card Interface Controller

Description

The ISO7816 Smart Card Interface IP core is a specialized UART, designed to meet specific ISO/IEC 7816-3 protocol requirements. Its flexibility can be tailored to meet T=0, T=1 and T=2 transmission protocols. The ISO7816 Smart Card Interface is designed to connect to the AMBA bus and can be interface to most peripheral expansion buses.

Features

  • Even/odd parity or no parity generation and detection
  • Programmable one or two stop bits per character
  • Asynchronous half and full duplex modes
  • Auto switching between transmit and receive modes
  • Error auto-detect and auto-retry
  • 3 maskable interrupts
  • Programmable baud rate clock
  • Transmit and receive FIFOs
  • Configurable receive and transmit buffer interrupts
  • Data format supported: 1 start bit, 8 data bits, optional parity bit, 1 or 2 stop bits
  • False start bit detection
  • Loop back and set break diagnostic modes
  • Compliant to AMBA 2.0 specifications

Applications

The ISO7816 Smart Card Interface provides a communication interface with the contacts of a smart card.  It may be used with an analog module for contacted (card inserted into reader) and contact-less (wireless) operation, or can be used stand-alone for a contacted interface only.

Smart Cards can be used for a variety of transactions, including credit/debit cards, access cards and “virtual shopping carts”.  They provide the same convenience of conventional magnetic-stripe cards with much higher data capacity, available security features, and capability to update information on the card.

The example above features all connections needed to integrate the Smart Card controller into a Smart Card reader SoC.  The CPU transfers data to and from the Smart Card via the APB interface.  Once the programmable FIFO threshold has been reached, the Smart Card controller interrupts the CPU.  If an analog interface controller is added to the system, two mode signals are provided to handshake between the two modules.

Deliverables

The IP core deliverable includes RTL code, synthesis scripts, test benches, and verification test suites.

Download

Download link coming soon

Real-Time Clock

Description

The RTC-0110 implements the functionality of a standard Real-Time Clock (RTC), and can be integrated into system on chip (SoC) designs. RTC is a collection of synchronous counters and comparators that allows the host processor access to the current time of day.  Alarms may be set and activated through the host firmware to start or stop various processes at different times and intervals.

The Real-Time Clock requires a separate 32.768Khz input clock.  The 32.768 KHz clock is divided down to generate a 1 Hz clock that drives the seconds register.  The other timer registers (minute, hour, day, etc.) update as appropriate on each 1 Hz clock pulse.

A separate timer that runs off of the 32.768 KHz clock is provided to generate a periodic interrupt to wake the processor when the primary system clock is disabled. This product can be integrated into SOCP-1010 Open-Silicon SoC Platform as a library component.

Features

  • Century/year, month, day, hour, minute and seconds registers
  • Day of week and day of year registers
  • Built-in daylight savings and leap year functions
  • Interrupt once, at a preset time
  • Recurring interrupts using compare masks
  • 32 KHz timer interrupt
  • Bit significant registers for day of week alarm and month alarm
  • Run/stop mode
  • Wait state on writes
  • Default time: midnight of Tuesday January 1, 1980
  • Compliant to AMBA 2.0 specifications

Applications

The RTC-0110 IP core can be integrated into any ASIC/SoC solution, and this solution can be used in various embedded systems to maintain the correct time of a system. A real-time clock runs of a special battery power which is different than the normal power supply of the system to keep track of the system time; even when systems are powered off. For example the RTC based ASIC/SoC can be used in digital video and still cameras, to maintain the time of the system, and print the current time on photographs or video.

Deliverables

The IP core deliverable includes RTL code, synthesis scripts, test benches, and verification test suites.

Download

Download link coming soon

16550-Compatible UART

Description

The UART-1011 IP core is functionally compatible with the 16450 and 16550 standard UART devices. The IP core functions as an asynchronous serial data input/output interface for system on chip (SoC) designs, performing serial/parallel conversion on data characters to and from a serial device. The UART-1011 includes a programmable baud rate generator and edge detector. The transmitter and receiver are each buffered with a 16 byte FIFO to reduce the CPU overhead. The IP is designed to connect to the AMBA APB bus or any on-chip peripheral bus. The data transfer is initiated via the AMBA APB interface. An interrupt is generated upon the completion of data transfer. It can be integrated into the SOCP-1010 Open-Silicon SoC Platform as a library component.

Features

  • Functionally compatible with 16450/16550
  • Programmable baud rate generator
  • Independent receive clock
  • 5-, 6-, 7-, or 8-bit characters
  • Even, odd or no parity
  • 1, 1.5, or 2 stop bits
  • False start bit detection
  • Line break generation and detection
  • Complete status reporting
  • Fully prioritized interrupt systemcontrol
  • Internal diagnostics, loopback, overrun,framing error and parity
  • Modem control functions (CTS, RTS, DSR, DTR, RI and DCD)
  • Easily interfaces to CPUs and DSPs
  • Available as library component for Open-Silicon SoC platform
  • Compliant to AMBA 2.0 specifications

Applications

The UART-1011 IP core is used to communicate between a computer’s interface and its serial devices. This includes terminals, modems or other peripheral devices such as scanners, mice, or keyboards that support asynchronous serial data transfers. It can also be used in custom applications such as handheld PCs. The UART modem control and status signals can be used for software flow control. It can also be used as a debug interface, in any SoC design. The above example features the IP core integrated into a SoC environment. The UART interface communicates to the PC through an RS-232 cable.

Deliverables

The IP core deliverable includes RTL code, synthesis scripts, test benches, and verification test suites.

Download

Download link coming soon

UART-Lite

Description

The URTL-1010 functions as an asynchronous serial data input/output interface for system on chip (SoC) designs, performing serial/parallel conversion on data characters to and from a serial device. The URTL-1010 IP core is derived from the standard UART device, and is optimized for low gate count while retaining the most commonly used functions.

Because of its small size relative to the 16550 UART, the URTL-1010 is ideal for system monitoring, self-test and debug.  The URTL-1010 is designed to connect to the AMBA bus and can interface to most peripheral expansion buses.

Features

  • Programmable baud generator
  • Independent receive clock
  • 8-bit characters
  • Even, odd, or no parity
  • 1 or 2 stop bits
  • False start bit detection
  • Line break generation
  • Complete status reporting
  • Independent transmit, receive, line status and data set interrupts
  • Internal diagnostics, loopback, overrun, framing error and parity
  • Optimized for standard test communication and low gate count
  • Easily interfaces to CPUs and DSPs
  • Available as library component for open-Silicon SoC platform
  • Compliant to AMBA 2.0 specifications

Applications

The UART-Lite IP core is used to communicate between a computer’s interface and serial devices such as terminals, scanners, pointing devices and keyboards that support asynchronous serial data transfers. It can also be used in custom applications such as handheld PCs. The UART-Lite is register-set compatible with the 16450 UART but without modem support, making it ideal for system monitoring and debug, and for devices which require standard asynchronous communication with a minimal die size.

The above example features the UART-Lite IP core integrated into an SoC environment. The UART interface communicates to the PC through an RS-232 cable. Direct control of the data transfer is initiated through the APB controller. An interrupt is asserted at the completion of each data transfer.

Deliverables

The IP core deliverable includes RTL code, synthesis scripts, test benches, and verification test suites.

Download

Download link coming soon

SmartUART

Description

The URTS-1012 functions as an asynchronous serial data input/output interface for system on chip (SoC) designs, performing serial/parallel conversion on data characters to and from a serial device. The URTS-1012 IP core is derived from the standard UART device, and is optimized for low gate count while retaining the most commonly used functions. This IP core is capable of autobaud/autoformat detection, and supports hardware/software data flow control.

Because of its automatic hardware/software flow control features and DMA capability, the URTS-1012 is ideal for communication systems.  The DMA interface enhances system performance by off-loading the processor when transferring large amounts of data.  The URTS-1012 is designed to connect to the AMBA bus and DMA channel interfaces, and can interface to most peripheral expansion buses.

Features

  • Programmable baud rate
  • Autobaud/autoformat detection
  • Automatic hardware/software flow control detection
  • 7- or 8-bit characters
  • Even, odd, stick or no parity
  • 1, 1.5 or 2 stop bits
  • Support for PIO or DMA interfaces
  • Independent transmits, receive, line status and data set interrupts
  • Internal diagnostics, loopback, overrun, framing error and parity
  • Optimized for standard test communication and low gate count
  • Easily interfaces to CPUs and DSPs
  • Available as library component for Open-Silicon SoC platform
  • Compatible with 16550 UART
  • Compliant to AMBA 2.0 specifications

Applications

The SmartUART IP core is used for communication between a computer and different peripheral devices. These would include serial devices such as terminals and modems, or other peripheral devices including scanners, mice and keyboards that support asynchronous serial data transfers. It can also be used in custom applications, such as handheld PCs. The SmartUART   modem control and status signals can be used for automated software/hardware flow control of data to and from system memory.  The combination of hardware support for flow control and DMA to memory is especially useful for transferring large blocks of data with minimal processor intervention.

The above example features the SmartUART IP core integrated into an SoC environment. The UART interface communicates to the PC through an RS-232 cable. Direct control of the data transfer is initiated through the CPU interface.  Data can be transferred via the CPU or directly to memory via the DMA interface.  An interrupt to the CPU is asserted at the completion of the data transfer.

Deliverables

The IP core deliverable includes RTL code, synthesis scripts, test benches, and verification test suites.

Download

Download link coming soon

Serial I2C Master

Description

The I2CM-1010 Serial I2C master is designed to support serial interfaces with various I2C-compliant devices. It allows firmware to configure the serial I2C interface to different modes to support a number of external serial peripheral devices. The I2C master is designed to support 8-bit devices. Address and device address width are programmable. Transfers can also be single-byte and burst transfers.

All transfer protocols required for I2C operation are performed automatically by hardware, including the two-way acknowledgment of data character receipt by the external device and the I2C master. The I2C master does not support multi-master arbitration or device wait.

Features

  • Even/odd parity or no parity generation and detection
  • Programmable one or two stop bits per character
  • Programmable address and device address widths
  • Asynchronous half and full duplex modes
  • Auto switching between transmit and receive modes
  • Error auto-detect and auto-retry
  • 3 maskable interrupts
  • Programmable baud rate clock
  • Transmit and receive FIFOs
  • Configurable receive and transmit buffer interrupts
  • Data format supported:
    • 1 start bit, 8 data bits, optional parity  bit, 1 or 2 stop bits
  • False start bit detection
  • Loop back and set break diagnostic modes
  • Available as library component for Open-Silicon SoC platform
  • Compliant to AMBA 2.0 specifications

Applications

The I2C interface provides support for a high data rate interface with a low pin count. The I2C master acts as an interface between an embedded CPU and external I2C peripheral devices. Such devices include flash memory, or other ASICs and ASSPs. The interface acts the interface master, initiating data transfers to and from the peripheral device.

The example above shows the I2C master core used to interface to serial flash. This application is commonly used to store device configuration data to a removable media card.  Example applications include cell phones and game controllers. The I2C master communicates to the flash through two bi-directional pins. The APB interface is used for CPU control of the I2C core. When data transfer is complete, an interrupt is generated to the interrupt controller; the CPU reads the status of the block and services the interrupt.

Deliverables

The IP core deliverable includes RTL code, synthesis scripts, test benches, and verification test suites.

Download

Download link coming soon

Serial I2C Master/Slave

Description

The I2MS-1011 is designed to support serial interfaces with various I2C-compliant hosts and devices. The I2MS-1011 can be configured to operate either in master or slave mode. The IP core is designed to support an 8-bit data bus with single or bursts transfers based on the data size configured in the control register. All transfer protocols required for I2C operation are performed automatically by hardware, including the two-way acknowledgment of data character receipt. The IP core supports multi-master arbitration, and the arbitration success or failure is indicated through a status bit.

The I2MS-1011 is designed to connect to AMBA bus, and it can interface to most peripheral expansion buses. The IP core can be integrated into the SOCP-1110 Open-Silicon SoC Platform as a library component.

 FEATURES

  • Supports the synchronous inter-integrated circuits (I2C) master and save protocol
  • Configurable to operate in master or slave mode
  • Generates start, stop, repeated start and ACK.
  • Detects start, stop, repeated start, and busy state
  • Supports multi-master arbitration
  • Programmable clock rate
  • Option to skip transmission of device address in-case of single host or device
  • Programmable device address width up to 8-bit
  • Programmable address width up to 8-bit
  • One-byte write and read buffer
  • Sequential (burst) byte-read and byte-write capability
  • Supports standard and fast mode
  • Compliant to AMBA 2.0 specifications

Applications

The I2C interface provides support for high data rate transfers with low pin count. The I2MS-1011 acts as an interface between embedded processors and external I2C peripheral devices. Such devices include EEPROM memories, ASICs and ASSPs. While operating in master mode, the I2MS-1011 initiates data transfers to and from the peripheral devices. Whereas, in slave mode, the I2MS-1011 responds to the master’s requests with acknowledgments and 8-bit data that is requested by master.

The above example shows the I2MS-1011 IP core being used to interface to a serial EEPROM. This type of application is commonly used to store device configuration data, and is commonly used in cell phones and game controllers.  The I2MS-1011 communicates to the EEPROM through the two bi-directional pins, and the CPU controls the I2MS-1011 through the MBA bus.  When data transfer is complete, an interrupt is generated to the interrupt controller; and the CPU services the
interrupt.

Deliverables

The IP core deliverable includes RTL code, synthesis scripts, test benches, and verification test suites.

Download

Download link coming soon

SPI Flash Controller

Description

The SPFL-1010 Serial Flash Controller is designed to support serial flash memories in a system. The flash controller supports the SPI modes 0 and 3, which are the most commonly used modes in flash devices. The flash controller translates the read command from the AMBA bus or any other on chip bus into a read command to the flash device, and provides the required data to the on-chip requestor.

The SPFL-1010 can be used to erase and reprogram the flash device. The CPU can program the required configuration and control registers in the controller through AMBA bus to execute an ERASE, PROGRAM or WRITE command to the device. The IP core will then generate the required cycles to the device. In addition any other control commands such as read ID, status, etc. can also be issued in the same manner.

The SPFL-1010 is designed to connect to AMBA or any other on chip buses, and can be integrated into the SOCP-1010 Open-Silicon SoC Platform as a library component.

 Features

  • Serial Peripheral Interface (SPI) compatible
  • Supports both SPI mode 0 and 3
  • Directly  memory mapped to SoC memory space
  • Programmable payload length
    • Opcode + address + data
  • Programmable interface clock rate
  • Supports following SPI flash instructions:
    • WREN
    • WRDI
    • RDSR
    • WRSR
    • READ
    • PROGRAM
    • SECTOR ERASE
    • CHIP ERASE
    • RDID
  • Compliant to AMBA 2.0 specifications

 

Applications

The SPFL-1010 can be integrated in ASIC/SoC/FPGA based solutions, to provide access to non-volatile memory in a system. This memory can then be used to store either system specific information, or to use it as boot memory. The use of a serial device reduces the pin count in a system, while providing the required performance. The SPFL-1010 offloads the CPU from control of data transfer to/from serial flash memory, which in turn increases the overall memory subsystem performance.

As shown in the above diagram the CPU is booting up from an off chip SPI flash memory while using the SPFL-1010.

Deliverables

The IP core deliverable includes RTL code, synthesis scripts, test benches, and verification test suites.

Download

Download link coming soon

Security IP

Open-Silicon solution for security IP consists of the AES IP and SHA engine. The AES is dedicated hardware accelerator that performs cipher/decipher functions on incoming data. The AES IP includes a integrated bus interface and DMA engine that allows for efficient implementation. SHA IP is an hardware accelerator to compute the HASH keys for different algorithms including MD-5, SHA-1, SHA-2.

AES Accelerator

Description

The AESE-1010 is a dedicated hardware accelerator that performs the cipher/decipher function on the incoming data. The AESE-1010 supports both PIO and DMA mode of operation. The AESE-1010 supports the three AES key sizes 128, 192, 256-bit, and performs offline key expansion. It supports AES ECB, CBC, CTR and CCM modes of operation and is fully compliant with IEEE 802.16e and ZigBee specifications for CTR and CCM modes.

The AESE-1010 is designed to interface to the host processor through the AMBA APB interface and to memory through a DMA channel.  It can interface to most peripheral expansion buses.

Features

  • 128-bit or 32-bit datapaths.
  • Programmable ECB, CBC, CTR and CCM support.
  • Programmable 128, 192, 256-bit AES key size support.
  • Key expansion and storage, built-in support for expanded key.
  • Fully compliant with FIPS 197 specifications.
  • Fully compliant with IEEE 802.16e and ZigBee specifications for CTR and CCM modes.
  • For AES CCM mode, support for Additional Authenticate Data (AAD) length from 0 to (216 -1) octets.
  • Full-duplex 2-channel DMA for transmitting and receiving data.
  • Scatter-gather operations supported.
  • Compliant to AMBA 2.0 specifications

Applications

The AES accelerator is used to encrypt and decrypt data. The AES accelerator is applicable to any system that needs to provide hardware acceleration for the purposes of encryption and decryption of data such as storage, 802.16 MAC layer, TCP/IP stacks, security and surveillance.

The example above shows the AES controller being used in a secure solid state storage device.

Deliverables

The IP core deliverable includes RTL code, synthesis scripts, test benches, and verification test suites.

Download

Download link coming soon

SHA Engine

Description

The HASH-1010 is a dedicated hardware accelerator that calculates the hashing function (or the message digest) for given byte oriented data or message. The HASH-1010 supports both PIO and DMA mode of operation. The HASH-1010 implements the most widely used hashing algorithms SHA-1, SHA-2(256) and MD5 in a single IP core. The support for the multiple modes makes the SHA  engine particularly suitable for implementation of internet security protocols, such as IPSec that requires multiple authentication algorithms. The input data is padded according the specific requirement of the hashing algorithm and endian-conversion is performed in the hardware.

The HASH-1010 is designed to interface to the host processor through the AMBA interface and to memory through a DMA channel.  It can interface to most peripheral expansion buses.

Features

  • SHA-1, SHA-2(256) and MD5
  • Compliant to FIPS 180-2 specifications
  • Compliant to MD5 RFC 1321 specifications
  • Message pre-processing, data framing and endian conversion done autonomously as per specification
  • Full Duplex, 2-channel DMA
  • Scatter-gather operations supported
  • Compliant to AMBA 2.0 specifications

Applications

SHA Engine functions such as SHA-1, SHA-2(256) and MD5 are extensively used in digital signatures, on-line banking, electronic fund transfer and electronic data transfer authentication applications.

The example above shows the SHA  engine being used in an IPsec offload hardware engine.

Deliverables

The IP core deliverable includes RTL code, synthesis scripts, test benches, and verification test suites.

Download

Download link coming soon

LZW Compressor and De-Compressor

Description

The LZW compressor encodes strings by reference to a dynamically generated dictionary. It exploits the redundancy of data for efficient compression by compressing uses up to 13 bit codewords. Each entry in the dictionary is a string and it is initialized with every possible byte by storing the byte in a 16384 entry two way skewed cache structure and as each entry is a 25 bit value it creates the hash table with 51200 bytes.

This is implemented as four 4K x 25 bit single port RAM along with a  two way skewed cache structure which also includes a pair banked RAM for emulating a two port RAM.

The de-compressor unit decompresses the incoming data stream by using a dictionary size of 8192 locations. It shares the dictionary RAM with compressor. It also uses the compressor’s data RAM to store character strings, and the compressor’s compressed data RAM to store string tracking information.

 Features

  • Greater than 320 MHz operation at 90 nm
  • 1 byte per clock cycle processing
  • 32 bit data paths
  • Compression enable/disable
  • Bypass option
  • Automatic compression abort when output compressed size is greater than the input data size
  • LZW compression/decompression
  • Variable length code words (up to 13 bits)
  • Compliant to AMBA 2.0 specifications

Applications

Some of the applications domain where LZW utilized is given below:

  • Used in storage such as SSD, disk drive and others
  • An optional part of the GIF and TIFF image file formats

Deliverables

The IP core deliverable includes RTL code, synthesis scripts, test benches, and verification test suites.

Download

Download link coming soon

Error Correction and Detection

Description

The Bose-Chaudhuri-Hochquenghem (BCH) Error Correcting Code (ECC) is a complete BCH Encoder and Decoder IP core that is implemented over GF(13).  The IP core includes two independent units, a parallel BCH Encoder for the generation of the BCH code and an independent BCH Decoder IP core based on Syndromer, Berlekamp-Massey, Chien Search and Error Correction units. The encoder and all the decoder units can be used separately or together.

The interface provided on the BCH Encoder & Decoder is generic and can be easily combined with any type of chip point-to-point buses. Open-Silicon also provides the AMBA bus interfaces that can be easily used with the BCH Encoder and BCH Decoder IP cores.

Features

  • 8-bit architecture
  • Parallel and highly optimized BCH Encoder
  • Parallel syndromer for BCH Decoder
  • Inversion-less BM for error locator polynomial with a latency of 2t cycles
  • Highly efficient and area optimized implementation for Chien search
    • Supports 16, 32 and 64-bit parallel factors
    • Can be customized to meet data rate and area requirements
  • Based over GF(13) and available on GF(14)
  • Programmable error correction values of (t) upto 55
    • Configurable to support any value for t
    • Configurable error correction polynomial
  • Multipliers are coded in the form of library so that they can easily be used/replaced and can provide excellent synthesis results
  • Usage of constant field multipliers as needed
  • Support 512 bytes of information (k)
    • Configurable to support any value for k
  • High throughput

Applications

BCH codes are used in wide range of applications from digital communications to storage, some of them are:

  • Satellite communications<
  • Storage devices
  • Digital television/DVB
  • Wireless or mobile communications such as WIFI, WIMAX
  • High-speed modems such as ADSL, xDSL, etc.
  • Power line standards

The above example features the BCH IP core integrated with a NAND Flash Controller to manage the inherent soft and random errors, as well as burst errors in MLC/SLC NAND Flash devices.  The BCH IP core can be used to detect and correct multiple errors.

Deliverables

The IP core deliverable includes RTL code, synthesis scripts, test benches, and verification test suites.

Download

Download link coming soon

I/O Controllers

I/O Controllers

Open-Silicon I/O controllers consists of storage controllers, ECC (error-correcting codes) and compression engines. The IP can be used for both solid state and rotating media. The IP offering is highly-configurable and can be customized to meet area/performance requirements. The IP blocks can be used as a stand-alone point solution or used as part of Open-Silicon customized and verified storage system on chip (SoC) solution.

Available I/O Controllers

SATA Device Controller

Description

The SATD-1110 is designed to interface between a host system and a storage controller, either solid state or rotating media.  The Serial ATA IP core decodes incoming host commands and sets up the proper interrupts and status for the local microprocessor to handle ATA and ATAPI commands.  Data transfer commands can be automated for full data transfer with minimal firmware support.  The IP core can be programmed for full automation or full firmware handling of each phase of command handling.  The SATD-1110 provides a FIFO interface to an external DMA controller for integration into systems with custom DMA functionality, such as an automated cache controller.

The SATD-1110 has two system interfaces: a control interface and an interface to an external DMA controller. The SATD-1110 product supports the AMBA bus interface. This IP core connects to an external DMA controller via a FIFO interface.

Features

  • Serial ATA version 1.0a
  • 150 MB/s transfer rates
  • Software-compatible with PATA
  • Supports ATA-2 through ATA-6
  • 28 or 48-bit LBA addressing
  • Auto-read and auto-write
  • Interface to external DMA controller
  • Compliant to AMBA 2.0 specifications

Applications

The Serial ATA target IP core connects a Serial ATA storage device to a host system, with command interpretation handled by the target IP core in conjunction with an embedded processor.

The example above shows the Serial ATA target IP core integrated into a solid-state drive.  The Serial ATA target connects to the host system via a connector.  Command processing is performed by interrupting the embedded processor; the processor parses the command and configures the various components to handle the command.  Data is transferred through the ATA target IP core into a data buffer via a DMA engine.  ECC is calculated for the data and it is stored to the media under software direction.  Data is read from the media, corrected via ECC if necessary, and transferred from the buffer to the host via the Serial ATA IP core.

Deliverables

The IP core deliverable includes RTL code, synthesis scripts, test benches, and verification test suites.

Download

Download link coming soon

SATA Host Controller

Description

The SATH-1111 Serial ATA host interface IP core provides an efficient and easy-to-use interface to Serial ATA (SATA) devices. The IP core implements transfer speeds of 150 MB/s, and emulates programmable I/O, multi-word DMA, and Ultra ATA modes of operation. The IP core can be configured as a primary or secondary IDE controller, and emulates master/slave operation for support of up to two devices. The IP core interface to the SOC includes a DMA controller to optimize data transfers to and from the IDE devices, and provides PIO access via shadow registers.  For ease of integration, the  SATH-1111 includes a register set that is compatible with the Intel chip set.

The SATH-1111 has two system interfaces: a control interface and an interface to an external DMA controller. The SATH-1111 product supports the AMBA bus interface. This IP core connects to an external DMA controller via a FIFO interface.

Features

  • 1.2 Gb/s and 1.5 Gb/s (150 MB/s)
  • Large (48-bit) LBA support
  • Supports up to 2 independent devices
  • Can be configured as primary and secondary IDE controller
  • Scatter-gather descriptor-based DMA  engine
  • Intel register set compatible
  • Synchronous DMA interface for data transfers
  • Compliant with Serial ATA 1.0
  • Compliant to AMBA 2.0 specifications

Applications

The SATA Host IP core is used to control Serial ATA drives. The Serial ATA Host is applicable to any system utilizing IDE/ATA protocol with Serial ATA drives for data storage including desktop computers, servers, set-top boxes and test equipment.

The example above shows the Serial ATA Host controller in a PCI I/O chip for use in a personal computer.  The application supports two floppy drives, two Serial ATA drives and two IDE/ATAPI drives.  The Serial ATA host controller is configured as a primary port controller; the IDE Host controller is configured as a secondary port controller. The IP core’s DMA channel interfaces are connected to the PCI bus via an arbiter.

Deliverables

The IP core deliverable includes RTL code, synthesis scripts, test benches, and verification test suites.

Download

Download link coming soon

ATA/ATAPI IDE Host Controller

Description

The ATAP-6010 is designed to interface between a host system and a storage controller, either solid state or rotating media.  The ATA block decodes an incoming host command and sets up the proper interrupts and status for the local microprocessor to handle various ATA commands.  Many commands can be automated for full data transfer with minimal firmware support.  Options also exist for full firmware handling of each phase of command handling.  The ATAP-6010 includes advanced features such as Ultra-ATA support and support for very large storage devices.

The ATAP-6010 is designed to interface to the local processor through the AMBA APB interface and to a data buffer memory through a separate DMA channel.

Features

  • Supports ATA protocol
  • Supports ATA-2 through ATA-6
  • Programmable I/O modes 0, 1, 2, 3 & 4
  • Multi-word DMA modes 0, 1 & 2
  • Synchronous Ultra ATA-33, -66, -100 and -133 modes 0, 1, 2, 3, 4, 5 & 6
  • 28 or 48-bit LBA addressing
  • Auto-read and auto-write
  • Compliant to AMBA 2.0 specifications

Applications

The ATA target IP core connects an IDE storage device to a host system, with command interpretation handled by the target IP core in conjunction with an embedded processor.

The example above shows the ATA target IP core integrated into a solid-state storage device – a flash card.  The ATA target connects to the host system via a connector.  Command processing is performed by interrupting the embedded processor; the processor parses the command and configures the various components to handle the command.  Data is transferred through the ATA target IP core into a data buffer via a DMA engine.  ECC is calculated for the data and it is stored to the media under software direction.  Data is read from the media, corrected via ECC if necessary, and transferred from the buffer to the host via the ATA IP core.

Deliverables

The IP core deliverable includes RTL code, synthesis scripts, test benches, and verification test suites.

Download

Download link coming soon

Ultra IDE Host Controller

Description

The UIDE-4010 IDE host interface IP core provides an efficient and easy-to-use interface to IDE and ATAPI devices. The IP core implements programmable I/O, multi-word DMA, and Ultra ATA-33, -66, -100 and -133 modes of operation and supports up to two devices. The IP core interface to the system on chip (SoC) provides PIO access and DMA capability to optimize data transfers to and from the IDE devices.  For ease of integration, the UIDE-4010 includes a register set compatible with the Intel chip set, including a descriptor-based scatter-gather DMA IP core.  This IP core is compatible with ATA-4 with Ultra ATA-33, -66, -100 and -133 extensions.

Features

  • Programmable I/O modes 0 – 4
  • Can be configured as primary and secondary IDE controller
  • Multi-word DMA modes 0 – 2
  • Synchronous Ultra ATA-33, -66, -100 and -133 modes 0 – 6
  • SpeedSelect allows core timings to be reprogrammed to support any ATA speed and timing mode at any clock frequency
  • Scatter-gather descriptor-based DMA engine
  • Intel PIIX register set compatible
  • Support for synchronous or asynchronous DMA interface for data transfers
  • Supports up to 2 devices with independent master/slave timing controls
  • Compliant to AMBA 2.0 specifications

Applications

The IDE Host block is used to control IDE disk drives. The IDE Host is applicable to any system utilizing IDE/ATA and ATAPI drives for data storage including notebook and desktop computers, servers, set-top boxes and test equipment. It is suitable for use with all form factor IDE/ATA and ATAPI drives.

The example above shows the IDE Host block used in a PCI I/O chip for use in a personal computer. The application supports four drives by instantiating two IDE Host controller blocks; one controller is configured as a primary port controller; the second controller is configured as a secondary port controller. The IP core includes a scatter-gather DMA channel compatible with the Intel scatter-gather DMA engine.

Deliverables

The IP core deliverable includes RTL code, synthesis scripts, test benches, and verification test suites.

Download

Download link coming soon

CE-ATA Host Controller

Description

The ATAH-1110 CE-ATA host controller IP core provides an efficient and easy-to-use interface to CE-ATA devices. The CE-ATA host controller provides both PIO and DMA mode for data transfer. The DMA controller can be used where high bandwidth and low CPU overhead is required. While designs that are targeted for low bandwidth applications, users can optionally remove the DMA controller and just use PIO mode thus saving on gate count. The host controller also provides a proprietary clock speed controlling feature which can be used to control data transfer rates dynamically when a CPU is not in a position to service regular data transfer rates.

The ATAH-1110 is designed to interface to the host processor through the AMBA APB interface and to memory through a DMA channel.  It can interface to most peripheral expansion buses.

Features

  • Compliant with multimedia card specification v4.1
  • Compliant with CE-ATA specification v1.1
  • Supports 1, 4, 8 bit data width MMC
  • ATA reduced command set implemented
  • CE-ATA task file emulation
  • Single channel DMA for transmitting and receiving data
  • Descriptor and non-descriptor mode
  • Compliant to AMBA 2.0 specifications

Applications

The CE-ATA Host block is used to control CE-ATA disk drives. The CE-ATA Host is applicable to any system utilizing CE-ATA drives for data storage, including notebook and desktop computers, set-top boxes, flash drives, cameras, MP3 players, security and surveillance.

The example above shows the CE-ATA Host block used in a PCI I/O chip for use in a personal computer.  A single DMA block and an APB Controller block provide access paths to the drives.  These blocks interface to a PCI block for connection to the host PCI bus.

Deliverables

The IP core deliverable includes RTL code, synthesis scripts, test benches, and verification test suites.

Download

Download link coming soon

CE-ATA Device Controller

Description

The ATAD-1110 CE-ATA device controller IP core provides an efficient and easy-to-use interface to CE-ATA Host. The CE-ATA device controller provides both PIO and DMA mode for data transfer. The DMA controller can be used where high bandwidth and low CPU overhead is required. While designs that are targeted for low bandwidth applications, users can optionally remove the DMA controller and just use PIO mode thus saving on gate count. The MMC device registers (OCR, CID, CSD, EXT_CSD, RCA, and DS) are implemented in hardware. The CE-ATA MMC command 60 and 61 are implemented.

The ATAD-1110 is designed to interface to the host processor through the AMBA APB interface and to memory through a DMA channel.  It can interface to most peripheral expansion buses.

Features

  • Compliant with multimedia card specification v4.1
  • Compliant with CE-ATA specification v1.1
  • Supports 1, 4, 8 bit data width MMC
  • MMC Device Class 0 command set implemented
  • ATA reduced command set implemented
  • Single channel DMA for transmitting and receiving data
  • Descriptor and non-descriptor mode
  • Compliant to AMBA 2.0 specifications

Applications

The CE-ATA target IP core connects a CE-ATA storage device to a host system, with command interpretation handled by the target IP core in conjunction with an embedded processor.

The example above shows the CE-ATA target IP core integrated into a solid-state drive.  The CE-ATA target connects to the host system via a connector. Command processing is performed by interrupting the embedded processor; the processor parses the command and configures the various components to handle the command.  Data is transferred through the ATA target IP core into a data buffer via a DMA engine.  Data is encrypted and ECC is calculated on the data and it is stored to the media under software direction.  Data is read from the media, corrected via ECC if necessary, decrypted and transferred from the buffer to the host via the CE-ATA IP core.

Deliverables

The IP core deliverable includes RTL code, synthesis scripts, test benches, and verification test suites.

Download

Download link coming soon

Hybrid Memory Cube Controller IP

Hybrid Memory Cube (HMC) represents an entirely new category of high-performance memory, delivering revolutionary performance and power in a dramatically reduced footprint.

Open Silicon’s HMC Controller IP provides the industry’s first, highest performance and most flexible solution for integrating the many benefits of HMC technology into next-generation systems.

Open-Silicon’s HMC Controller IP is a high-performance, flexible soft macro implementation of the Micron HMC Generation 2 Interface Protocol. The design provides system designers with the fastest, lowest risk solution for interfacing to the Hybrid Memory Cube.

The Open-Silicon HMC IP supports the Micron HMC Generation 2 Interface. As one of the developer members of the HMC Consortium, Open-Silicon plays a key role in developing the HMC Interface Specification and roadmap, as well as enabling industry application and ecosystem development.

Designed and tested to be easily synthesizable into 45nm – 22nm ASIC technologies, Open-Silicon’s HMC IP Core was uniquely built to seamlessly interface with SerDes from leading technology vendors. This allows Open-Silicon customers to quickly integrate the HMC IP Core into their process technology and SerDes vendor of choice.

Features

  • Transaction, link, and logical sub-block of the physical layer interface to 3rd party SerDes IP
  • Native interface or future AXI 4.0 option
  • Parameterizable number of native interface ports (1, 2 , 4 or 5)
  • Native interface ports have separate 256b read and write paths
  • User defined sideband information
  • Support for 10Gbps, 12.5Gbps, and 15Gbps SerDes speeds
  • Supports 16, 32, 48, 64, 80, 96, 112, and 128 byte requests
  • Mode read/write for configuration and status
  • Atomic commands
  • Power on initialization
  • Scrambling and descrambling
  • Error detection and automatic retry
  • Configurable error injection mechanisms for testability
  • Token based flow control
  • Tag management and tracking
  • Fully synchronous design
  • 625MHz – 937.5MHz

HMC Technology

Some of the prime benefits of HMC Technology as compared to other memory technologies are:

  • Provides over 15x the performance of a DDR3 module
  • Utilizes 70% less energy per bit than DDR3 DRAMs
  • 90% less space than today’s RDIMMs
  • The interface to the HMC from the host device uses a packet-based interface and high-speed SerDes links.

About the Consortium

The HMC Consortium is a working group made up of industry leaders who build, design-in, or enable Hybrid Memory Cube (HMC) memory technology. The group works to innovate and expand the capabilities of the next generation of memory-based. For more information, please visit: http://www.hybridmemorycube.org/

Deliverables

  • Synthesizable RTL
  • Example CAD scripts for synthesis, LEC, and linting
  • Assertions for the native interface and config registers
  • Testbench, bus functional models, and interface monitor
  • Documentation
    • Open-Silicon IP specification
    • Configuration and status register manual
    • Testbench, bus function model (BFM), and monitor documentation

Download

Download link coming soon

Ethernet IP

The Open-Silicon Ethernet IP solutions are highly modular enabling system level solutions from 10G to 100G, and are ideal for low-latency applications such as data center switches where higher performanceis required. The small footprint implementation leads to lower gate counts and smaller die sizes, which allow for an efficient design in high-port count devices. As a highly configurable design that can support a number of different SerDes configurations and user interface options the 40G/100G MAC and PCS offers easier trade-off analysis and quick tuning of the IP to meet user requirements. Comprehensive system-level verification and automated infrastructure, with automated web-based IP builder, allows quick trade-off analysis and system-level integration.

40G Physical Coding Sublayer IP

Description

The 40Gb PCS is a 40Gb/s Physical Coding Sublayer (PCS) to be used in high port density switching ASICs. The 40Gb PCS interfaces to four 10G SERDES as well as a MAC and provides PCS functionality with low gate count thus providing both area and power savings. The 40Gb PCS is compliant with 802.3ba clause 82 for 40Gbps (802.3-2008 clause 49 specifies 10Gb PCS of which 40G is a natural extension) operation. The 40G PCS is architected for both ASIC and FPGA implementations.

Interface Definitions

  • MAC to PCS: The MAC facing interface is 128 bits wide, 16 control bits and is either a double wide XLGMII bus, or a standard width XLGMII bus if implemented as 64b data path. To meet line rate requirements, the bus speed must be 322.215MHz (644.43MHz with 64b data path)
  • SerDes to PCS: The SerDes facing interface is configured to match common SerDes speeds and widths of 64-bit @ 161.10 MHz, 32-bit @ 322.215 MHz or 16-bit @ 644.43 MHz. There are 4 such interfaces which carry a single parallel representation of a 40Gb stream. On TX, all SerDes can be assumed to use the same clock. On RX, 4 separate clocks are received and may differ somewhat. An asynchronous fifo is implemented per lane inside the deskew block to adapt the rates to the core clock.

Features

  •  PCS Architecture to support the following speed: 1x40Gbps
  • Interfaces to four SERDES on line side. SERDES operating at 10.312Gbps for 40Gbps operation
  • Implements the following PCS Configurations. 1x40GBASE-R PCS while operating at 40Gbps
  • Provides 128-bit data, 16 bit control XLGMII like interface the MAC. Alternatively provides 64-bit data, 8 bit control XLGMII interface for narrower data path
  • Implements the PCS Register set(s) for operation and control of the PCS modules
  • Capable of interfacing to SERDES Modules with a 64-bit or 32-bit or 16-bit wide data buses

40G PCS Functions

  • Implements the full register set as specified in the clause 45 to report status and enable the control of the 40G PCS
  • Performs encoding of two 64-bit MAC data + 8-bit MAC control into two 66-bit code word. (In 64 bit data path configuration, a single encoding is performed)
  • Scrambling of the resultant 66-bit code words
  • Performs block synchronization of the receive data stream to determine 66-bit code groups
  • Descrambling of data and decoding of code groups
  • Performs insertion and deletion of alignment markers in the transmit/receive streams
  • Implements detection and reporting of fault conditions and code synchronization errors
  • Implements lane deskew and reordering functions
  • Implements optional Test pattern generator and checker as defined in the specification
  • Implements Bit Error Rate(BER) monitoring with high error rate indication, thus providing constant line quality monitoring capability
  • Continuously monitors the receive data stream for BER monitoring
  • Link status reporting for fault conditions

Target Applications

  • High-port density switches
  • Aggregation devices
  • Network Interface (NIC) Card

Deliverables

  •  Data Sheet
  • Integration Guide
  • Programmers Guide
  • RTL database
  • Integration test suite
  • ASIC deliverables (Synthesis scripts, timing constraints, DFT guidelines)

Download

Download link coming soon

40G Ethernet Media Access Controller IP

Description

The 40G Ethernet Media Access Controller (MAC) implements essential protocol requirements for operation of 40 Gbp/s Ethernet (IEEE 802.3ba) compliant node. The 40GMAC IP Core supports configurable FIFO’s on both transmit and receive side to handle application’s latency during the frame transmission and reception. The 40GMAC is compliant to IEEE 802.x standard for full-duplex operations by supporting PAUSE frame and PAUSE operations. The 40G MAC is architected for both ASIC and FPGA implementations using a standard tool flow.

Interface Definitions

  • FIFO: Flexible FIFO interface with start of packet, end of packet and additional control signals. Interface can operate on an independent clock from core’s transmit clock.
  • XGLMII: The 40GMAC Core interfaces to the 40G PCS modules using the IEEE802.3 compliant XLGMII with 64-bit data, 8-bit control or 128-bit data, 16-bit control.
  • Registers Interface: Simple interface to read/write control and configuration registers

Features

  • Operates at independent transmit and receive 625 MHz clocks as defined in clause 81 of IEEE 802.3ba specification
  • Q-Tag frame support. Accounts for VLAN tags while checking for the max. frame lenght
  • Full-duplex mode operation while supporting PAUSE frame based flow control
  • 64-bit Counters for Statistics
  • Operates at independent transmit and receive 625 MHz clocks as defined in the Clause 81 of the IEEE 802.3ba specification
  • Generation of Clause22 (Direct) or Clause45 (Indirect) compliant management frames under software control on MDC/MDIO interface to talk to external PHY device
  • 802.3 Compliant MIB, SNMP, RMON management support by using variety of 64-bit counters
  • Configurable Transmit and Receive FIFO’s
  • Support’s jumbo frames during both transmit and receive operations
  • Provides a simple registers interface to read/write registers for control and configuration of the 40GMAC

40G MAC Transmit Functions

  • Variable length (96 BT, 128 BT, 256 BT) inter frame gap (IFG) on back to back frame transmission, with default value of 96 Bit Times of IFG
  • Deficit IDLE Counter to maintain an average IFG of 96 bit times
  • Automatic generation of FCS/PAD during transmission on per frame basis
  • Option to disable PAD and/or CRC32 insertion on transmission on a per frame basis

40G MAC Receive Funtions

  •  Preamble detection and stripping on reception
  • Handles minimum IFG of 5 bytes during back to back frame reception
  • Automatic checking of the FCS field for correct CRC value
  • Automatic checking of Runt frames and option to filter them out
  • Automatic checking of the DATA field length in case of 802.3 type frames with length field
  • Configurable field to detect MaxFrameLen frames
  • Automatic adjustment to the MaxFrameLen field for VLAN Tagged frames
  • Detection of receive error indication on PCS bad code word decodes
  • Comprehensive 64-bit status information provided on each receive frame

Flow Control

  •  PAUSE control frame generation with programmable pause quanta
  • Ability to generate PAUSE control frames on FIFO Almost-Full and FIFO Almost-Empty conditions
  • Automatic detection of PAUSE frames with DA field of either the reserved multicast address or the unicast MAC Address(s) of the device
  • Ability to block the PAUSE frames received
  • Support for priority based flow control as defined in the IEEE 802.1Qbb specification.

Target Applications

  •  High-port density switches
  • Aggregation devices
  • Network Interface (NIC) Card

Deliverables

  •  Data Sheet
  • Integration Guide
  • Programmers Guide
  • RTL database
  • Integration test suite
  • ASIC deliverables (Synthesis Scripts, Timing Constraints, DFT Guidelines)

Download

Download link coming soon

100G Physical Coding Sublayer IP

Description

The 100G PCS is a 100 Gb/s Physical Coding Sublayer (PCS) to be used in high port density switching ASICs. The 100Gb PCS interfaces to ten 10G SerDes as well as MAC and provides PCS functionality with low gate count thus providing area and power savings. The 100Gb PCS is compliant to 802.3ba clause 82 for 100 Gbps (802.3-2008 clause 49 specifies 10Gb PCS of which 100G is a natural extension) operation. The 100G PCS is architected for both ASIC and FPGA implementation using a standard tool flow.

Interface Definitions

  • MAC to PCS: The MAC facing interface is 20 bits data, 40 control bits or 256 bit data,32 bit control or 128 bit data, 16 bit control.
  • SerDes to PCS: The SerDes facing interface is configured to match common SerDes speeds and widths of 64-bit @ 161.10 MHz, 32-bit @ 322.215 MHz, 20-bit @515.625 MHz or 16-bit @ 644.43 MHz. There are 10 such interfaces which carry a single parallel representation of a 100Gb stream. On TX, all SerDes can be assumed to use the same clock. On RX, 10 separate clocks are received and may differ. An asynchronous fifo is implemented per lane inside the deskew block to adapt the rates to the core clock.

Features

  • PCS Architecture to support the following speed: 1x100Gbps
  • Interfaces to ten SERDES on line side. SERDES operating at 10.312Gbps for 100Gbps operation
  • Implements the following PCS Configurations. 1x100GBASE-R PCS while operating at 100Gbps
  • Flexible clocking options
  • Provides 320-bit data, 40 bit control (optional 256b/32b or 128b/16b) CGMII like interface to the MAC
  • Implements the PCS Register set(s) for operation and control of the PCS modules
  • Capable of interfacing to SERDES Modules with a 64-bit or 32-bit or 16-bit wide data buses
  • Easily contrallable via simple register interface
  • Integrated test pattern/ PRBS generator and checker
  • Low gate count for area and power savings

100G PCS Functions

  • Implements the full register set as specified in the clause 45 to report status and enable the control of the 100G PCS
  • Performs encoding of two 64-bit MAC data + 8-bit MAC control into two 66-bit code word. (In 64 bit data path configuration, a single encoding is performed)
  • Scrambling of the resultant 66-bit code words
  • Performs block synchronization of the receive data stream to determine 66-bit code groups
  • Descrambling of data and decoding of code groups
  • Performs insertion and deletion of alignment markers in the transmit/receive streams
  • Implements detection and reporting of fault conditions and code synchronization errors
  • Implements lane deskew and reordering functions
  • Implements optional Test pattern generator and checker as defined in the specification
  • Implements Bit Error Rate(BER) monitoring with high error rate indication, thus providing constant line quality monitoring capability
  • Continuously monitors the receive data stream for BER monitoring
  • Link status reporting for fault conditions

Target Applications

  •  High-port density switches
  • Aggregation devices
  • Network Interface (NIC) Card

Deliverables

  •  Data Sheet
  • Integration Guide
  • Programmers Guide
  • RTL database
  • Integration test suite
  • ASIC deliverables (Synthesis scripts, timing constraints, DFT guidelines)

Download

Download link coming soon

100G Ethernet Media Access Controller IP

Description

The 100G Ethernet Media Access Controller (MAC) implements essential protocol requirements for operation of 100 Gbp/s Ethernet (IEEE 802.3ba) compliant node. The 100GMAC IP Core supports configurable FIFO’s on both transmit and receive side to handle application’s latency during the frame transmission and reception. The 100GMAC is compliant to IEEE 802.x standard for full-duplex operations by supporting PAUSE frame and PAUSE operations. The 100G MAC is architected for both ASIC and FPGA implementations using a standard tool flow.

Interface Definitions

  • FIFO: Flexible FIFO interface with start of packet, end of packet and additional control signals. Interface can operate on an independent clock from core’s transmit clock.
  • CGMII_128/CGMII_256/CGMII_320: The 100GMAC Core interfaces to the 100G PCS modules using the IEEE802.3 compliant CGMII with:
    • 128-bit data, 16-bit control and independent receive and transmit clocks running at 781.25 MHz Clock.
    • 256-bit data, 32-bit control and independent receive and transmit clocks running at 390.625 MHz Clock.
    • 320-bit data, 40-bit control and independent receive and transmit clocks running at 312.5MHz
  • Registers Interface: Simple interface to read/write and control the configuration registers

Features

  • 100Gbps IEEE 802.3ba Compliant MAC
  • Support for IEEE 802.3 (2008)
  • Small footprint design for low area and power savings
  • Optimized for ASIC implementations
  • Configurable Data path (128/256/320 bits)
  • IEEE-802.3ba compliant CGMII like interface (Clause 81) with 128 bit data, 16 bit control or 256 bit data, 32 bit control or 320-bit data, 40-bit control to interface to 100G PCS Module
  • Interfaces with application side with a Flexible FIFO interface with control for easy integration
  • PAUSE Flow Control and Priority Pause Frames
  • 64-bit Counters for Statistics
  • Up to 3 VLAN Q-Tag Frame support. Accounts for the VLAN tags whilechecking for the Max Frame Length.
  • Full-Duplex mode of operation while supporting PAUSE frame based flow control.
  • 802.3 Compliant MIB, SNMP, RMON management support by using variety of 64-bit counters.
  • Configurable Transmit and Receive FIFO’s
  • Supports Jumbo Frames during both transmit and receive operations
  • Provides a simple Registers Interface to read/write registers for control and configuration of the 100GMAC
  • FPGA support available

10G MAC Transmit Functions

  • Variable length (96 BT, 128 BT, 256 BT) inter frame gap (IFG) on back to back frame transmission, with default value of 96 Bit Times of IFG Deficit IDLE Counter to maintain an average IFG of 96 bit times
  • Automatic generation of FCS/PAD during transmission on per frame basis
  • Option to disable PAD and/or CRC32 insertion on transmission on a per frame basis

100G MAC Receive Funtions

  • Preamble detection and stripping on reception Handles minimum IFG of 5 bytes during back to back frame reception
  • Automatic checking of the FCS field for correct CRC value
  • Automatic checking of Runt frames and option to filter them out
  • Automatic checking of the DATA field length in case of 802.3 type frames with length field
  • Configurable field to detect MaxFrameLen frames
  • Automatic adjustment to the MaxFrameLen field for VLAN Tagged frames
  • Detection of receive error indication on PCS bad code word decodes
  • Comprehensive 64-bit status information provided on each receive frame

Flow Control

  • PAUSE control frame generation with programmable pause quanta
  • Ability to generate PAUSE control frames on FIFO Almost-Full and FIFO Almost-Empty conditions
  • Automatic detection of PAUSE frames with DA field of either the reserved multicast address or the unicast MAC Address(s) of the device
  • Ability to block the PAUSE frames received
  • Support for priority based flow control as defined in the IEEE 802.1Qbb specification.

Target Application

  •  High-port density switches
  • Aggregation devices
  • Network Interface (NIC) Card

Deliverables

  • Data Sheet
  • Integration Guide
  • Programmers Guide
  • RTL database
  • Integration test suite
  • ASIC deliverables (Synthesis Scripts, Timing Constraints, DFT Guidelines)

 Download

Download link coming soon

DDR3 Memory Controller

Description

The DDRC-3010 is a high performance memory controller that provides a highly configurable interface to external Double Data Rate Synchronous DRAM (DDR3 DRAM). It can support up to 8 AXI User ports with an intelligent scheduler that is optimized for very high throughput. It also includes a CPU port that is optimized for low latency and timing critical transactions.  DDRC-3010 can support data rates up to 2133 Mbps.

The DDR3 memory controller can support a wide variety of memory devices and organizations including standard DDR3 DIMMs like UDIMM, RDIMM, etc.   The memory controller supports all the memory functions including automated initialization, self-refresh, auto-refresh and power-down etc.  The various DDR3 timing parameters are programmable. It can be configured to optionally support ECC, command /address parity and 1T/2T timing.

The DDR3 memory controller has a highly intelligent scheduler that includes multiple algorithms to improve the memory throughput.  The scheduler improves the memory throughput by reordering requests and minimizing the effects of various DDR internal operations like precharge/refresh, etc. The controller can support BL8 (burst length) or BC4 (burst chop) on-the-fly to improve memory throughput.

The memory controller supports DDR PHY Interface (DFI v2.1) to ease the integration of DDR PHY.  It supports the DFI frequency ratios of 1:2 and 1:4.

Features

  • Up to 2133 Mbps operation
  • Compliant to JEDEC DDR3 (JESD79-3E) standards
  • Support for standard DDR3 DIMMs
    • UDIMM including SODIMM
    • RDIMM including SSTE32882 parity
    • Optional ECC generation/checking
  • Support for x4, x8 and x16 memories
  • Support for up to 8 ranks (CS#)
  • Support for 1T and 2T timing
  • On-the-fly support for BL8 and BC4
  • Low latency (3 cycles or less)
  • DFI v2.1 support
    • Support for 1:1 and 1:2 frequency ratio
    • DFI timing parameters are highly-configurable
  • Highly-configurable DDR parameters
  • Page size, bank-size, device configuration etc.
  • DDR bus width support: 32b/64b
  • SDRAM self-refresh, auto-refresh and automated SDRAM initialization
  • Programmable DDR timing parameters
  • Power-down and self-refresh
  • FPGA support available

Applications

The DDR3 controller is highly configurable solution and can be targeted towards a wide variety of applications such as:

  • Enterprise applications such as communications and networking
  • High-Performance Computing
  • Mobile Applications
  • Consumer
  • Data Processing

Deliverables

  • Synthesizable RTL code
  • Synthesis and STA scripts
  • Comprehensive documentation
  • Testbench/Verification suites

Download

Download link coming soon