Designed to work with a wide variety of SPI bus variants, the core supports run-time control of several SPI protocol parameters. For example, the SPI frame width can be 1 to 4 bytes, the
most significant bit position in a frame, serial clock phase and polarity are all software- programmable. In master mode the core can control up to 32 slaves. A software controllable clock generator derives the serial clock for master mode, by dividing the frequency of a clock line dedicated for that purpose
The MAC-1G/MAC is a synthesizable HDL core of a high-speed LAN controller. It implements Carrier Sense Multiple Access with Collision Detection (CSMA/CD) algorithms defined by the IEEE 802.3 standard for media access control over the 10Mbps, 100Mbps and 1Gbps Ethernet. Communication with an external host is implemented via a set of Control and Status Registers and the DMA controller for external shared RAM memory. For data transfers the MAC-1G/MAC operates as
a DMA master. It automatically fetches from transmit data buffers and stores receive data buffers into external RAM with minimum CPU intervention. The linked list management enables the use of various memory allocation schemes. There is an interface for external dual port RAMs serving as configurable FIFO memories and there are separate memories for transmit and receive processes. Using the FIFOs additionally isolates the MAC-1G/MAC from an external host and provides resolution in case of latency of an external bus.
Network Interface Cards (NICs)
Routers, switching hubs
Double Data Rate 4 (DDR4) SDRAM Controller Core is designed for use in applications requiring high memory throughput, high clock rates and full programmability.
The core uses bank management modules to monitor the status of each SDRAM bank. Banks are only opened or closed when necessary, minimizing access delays. Up to 32 banks can be managed at one time.
The core supports all new DDR4 features, including: 3DS device configurations, write CRC, data bus inversion (DBI), fine granu-larity refresh, additive latency, per-DRAM addressability, and temperature controlled refresh.
FLASH memory controller ideal for interfacing to a wide range of parallel FLASH memory components . Features a fully synchronous command interface and a set of configurable timing parameters for compatibility with different devices.
Any application where non-volatile storage is required
Offline storage of parameters and data via your Chip
HEART (High Efficient Accumulative Repairing Technical) is a built-in self-repair (BISR) mechanism which uses to recover errors detected after memory testing and to improve yield rate. This mechanism is implemented with spare memories and a built-in redundancy analyze (BIRA) logics which is designed to allocate the redundancy. It needs a storable device (eFuse, OTP or registers) to store testing results after analysis.
We provides an efficient accumulative repairing solution to combine advantages of soft BISR mechanism and hard BISR mechanism for improving yield rate.
HEART can efficient repair faulty SRAM after using BRAINS. SoCs can mantain correctness of functions and avoid fatal error of system reault in SRAM's defect through SRAM's repairing technical.
HEART is SRAM accumulative repairing technical, and it combines advantages of Soft-repair and Hard-repair. HEART supports internal registers of SoCs and external storages of SoCs to record SRAM's faulty information. Once SoCs have new SRAM's defect after using them for a long time, users can repeated repair SRAM's defect through HEART. In addtion, HEART also support "On-Demad" testing and repairing requirement. It means that users can enable system registers of SoCs or signal of HEART to test and repair SRAM at one when SoCs have fatal error situations.
With improvement of technology node and IC design is geting more complex, the ratio of embedded memory in SoCs have been exceeding 50%. The fault types of memory are getting complex. The Memory BIST (Built-In Self-Test) is generated for efficient controlling IC cost. The traditional BIST method is inserted along with single memory. If there are many memories in SoCs, the area and testing time of SoCs are expanded a lot due to insertion of BIST. Therefore the SoCs' cost will increase rapidly because memory testing time is too long.
We devoted in developing SRAM testing solutions for a long time. BRAINS is based on memory testing patents to reduce testing time and increase yield rate. In addition, BRAINS has many unique features to increase SoCs' reliability and stability.