Interface IP
Interface IP cores support a wide array of interfaces, including Automotive IP, Chiplet and D2D IP, IP Stack IP, Interconnect IP, Memory Interface IP, Telco/OTN IP, and Video Interface IP. Automotive IP cores ensure reliable connectivity in vehicle systems, while Chiplet and D2D IP support the optimization of multi-chip designs for better performance. IP Stack IP enables seamless networking protocols, and Interconnect IP facilitates high-speed data transfer between system components. Memory Interface IP cores optimize access to storage and memory, and Telco/OTN IP supports high-performance telecommunications and optical networks. Video Interface IP cores provide seamless video data transmission, ensuring high-quality multimedia experiences.
-
PCIe 6.2 Switch
- 1 upstream port, up to 7 downstream ports
- Up to 128 lanes
- PCIe TLP routing: Configuration, Memory Write/Read, I/O and Messages Packets
- L1 and wake-up events forwarding
-
HBM4 Memory Controller
- Supports HBM4 memory devices
- Supports all standard HBM4 channel densities (up to 32 Gb)
- Supports up to 10 Gbps/pin
- Refresh Management (RFM) support
- Maximize memory bandwidth and minimizes latency via Look Ahead command processing
- Integrated Reorder functionality
-
PCIe 7.0 Switch
- Configurable from PCIe 7.0 x8/ PCIe6x16 @1GHz clock down to PCIe 5.0 x1
- Highly scalable with up to 31 configurable external or embedded endpoints
- Configurable Egress Buffer for non-blocking output queueing switch performance
- Flit mode to non-Flit mode conversion
- Low power optimized
- Superior performance through a nonblocking architecture
- Minimized footprint
-
PCIe 7.0 Retimer Controller with CXL Support
- Supports PCIe 7.0 128 GT/s speeds at up to x16 lanes
- CXL 3.0 aware
- Supports PIPE 6.2.1 compatible PHYs
- Optimized for low latency
- Highly-configurable equalization algorithms and adaptive behaviors
-
PCIe 7.0 Controller with AXI
- Optimized for high-bandwidth efficiency at data rates up to 128 GT/s
- Separate native TX/RX data path separating posted/Non posted/completion traffic
- Handles up to 4 TLPs per cycle
- Advanced PIPE modes and port bifurcation
-
PCIe 7.0 Controller
- Optimized for high-bandwidth efficiency at data rates up to 128 GT/s
- Separate native TX/RX data path separating posted/Non posted/completion traffic
- Handles up to 4 TLPs per cycle
- Advanced PIPE modes and port bifurcation
-
LPDDR5T / LPDDR5X / LPDDR5 Controller
- Support for all LPDDR5T/5X/5 devices
- Bank management logic monitors status of each bank
- Queue-based user interface with reordering scheduler
- Look-ahead activate, precharge, and auto-precharge logic
- Parity protection for all stored control registers
- PHY interface based on DFI 5.1 standard
-
Ethernet Enterprise Switch/Router IP Core - Efficient and Massively Customizable
- Full wire-speed on all ports and all Ethernet frame sizes.
- Store and forward shared memory architecture.
- Support for jumbo packets up to 32739 bytes.
-
GDDR7 Memory Controller
- Supports up to 40 Gbps per pin operation
- 2.5 GHz CK4 clock
- 1.25 GHz controller clock
- Internal data path 32x memory width (i.e. 256 bits for 8-bit memory)
- Optimized for high efficiency and low latency across a wide range of traffic scenarios (random/sequential, short/long bursts, etc.)
- Optimized command sequence for highest bus utilization including per-bank refresh scheduling: single queue structure handles look-ahead activates/ precharges and read/write ordering for minimal latency
-
CXL 3.1 Controller
- Ultra-low Transmit and Receive latency
- Internal data path size automatically scales up or down (256, 512 or 1024 bits) based on max. link speed and width for optimal throughput
- Supports backwards compatibility to PCIe 6.1
- Optional MSI/MSI-X register remapping to memory for reduced gate count when SR-IOV is implemented
- Optional QuickBoot mode allows for up to 4x faster link training, cutting system-level simulation time by 20%
- Loopback Mode support at DLL for CXL.mem and CXL.cache protocols
- Merged Replay and Transmit buffer enables lower memory footprint